CN1532810A - Audio frequency decoding device - Google Patents

Abstract

An audio decoding device is provided for decoding N<SUB>A </SUB>(where N<SUB>A</SUB>>1) channels of audio signals by a sub-band synthesis operation using sub-band synthesis filter data and sub-band signal data. The decoding device includes a first memory section for storing M<SUB>A </SUB>(where M<SUB>A</SUB><N<SUB>A</SUB>) channels of the sub-band synthesis filter data and the sub-band signal data, a second memory section for storing at least some of N<SUB>A </SUB>channels, an operation section for receiving encoded audio data and decoding the encoded audio data into sub-band signal data, and a data transfer section for, switching, by M<SUB>A </SUB>channels, the sub-band synthesis filter data and the sub-band signal data in the first memory section and the second memory section.

Description

Audio frequency decoding device

The application be that May 15, application number in 1998 are 98103250.8 the applying date, name is called the dividing an application of Chinese patent application of " compressed code decoding device and audio frequency decoding device ".

Technical field

Thereby the present invention relates to a kind of sound signal decoding equipment that is used for obtaining the sound signal of decoding according to the coding audio data of deciphering a plurality of passages such as MPEG 2 standards.

Background technology

In recent years, in video and field of audio devices, CD player and VTR machine are very universal, thereby more and more need to use large screen display and the pictures and sounds true to nature of regenerating is used for family's music appreciating and film.For satisfying this demand, the medium that carry video and audio signal have been developed, for example DVD etc.Simultaneously, in order in single medium, to store a large amount of video and audio signals, developed the method for store compressed signal on this medium.For example, the existing method that is used for the store compressed sound signal comprises MPEG Audio method and Dolby AC-3 method at present.

In order to obtain this compressed signal, sound signal is divided into a plurality of " pieces " with certain hour length, and signal is compressed by block.Then, another is referred to as the signal element of " frame " block of some to be combined formation.In decoding is handled, provide for each signal frame to be used for distinguishing mutually the auxiliary data of signal frame.Decoding equipment usually utilizes signal frame to carry out decoding and handles.

And a plurality of signal frames combine and form one " bag ".Each signal packet includes the time data of each signal frame that is used to refer in the signal packet, represents the decoded signal of first block in the regenerated signal frame when.According to time data, decoding equipment is externally exported the decoded signal of obtaining by decoding compressed signal block signal.Time data is mainly used in the video and audio signal of regeneration phase mutually synchronization.

To narrate conventional decoding compressed code equipment hereinafter with reference to the accompanying drawings.Figure 44 is the block diagram of the exemplary configuration of the explanation conventional equipment that is used for deciphering compressed code.

In Figure 44, decoding scheme 402 receives the compressed audio sign indicating number of importing with frame by input terminal 401, and it is deciphered.Output buffer 403 temporarily stores the sound signal of decoding.Output circuit 404 is externally exported the sound signal that is stored in the output buffer 403.

According to this configuration, compressed code inputs to input terminal 401 at first a frame one frame.Why compressed code is imported by a frame one frame ground is because according to the method described above signal frame is processed into the frame signal that can distinguish each other.The corresponding to interpretation method of method that decoding scheme 402 utilizes and signal compression is used is deciphered the code that is included in the frame signal, thereby produces a sound signal.Sound signal temporarily is stored in the output buffer 403.When having deciphered a signal frame, by output circuit 404 output audio signal externally.At this moment, the next frame signal code is inputed to input terminal 401, and decipher, write then in the output buffer 403 by decoding scheme 402.Like this, repeat this processing and decipher compressed code.

Yet this conventional equipment can not regeneration and vision signal and be used as the sound signal of other signal Synchronization such as timing signal of regeneration reference time, and only can be with different time output audio signals.Also having the another one problem is that the storage volume of requirement output buffer will have a frame signal big like that.

And, in order to calculate the decoding coded signal, need with storer that can zero access as temporary transient storage will calculated coded signal storer.Such storer is very expensive, and strong request reduces the storage volume of required storer.

For example, below in conjunction with MPEG2 the necessity of this high speed access storage is described in detail.

According to audio frequency MPEG algorithm, utilizing before the subrane analysis filterbank is compressed into the frequency band of 32 subrane signals with wave band, encoding device is divided sound signal.For example, by one 512 tap multiphase filter (PFB), go on foot division of signal wavelet segment signal by following (1)-(3):

(1) to 512 sample X0 of input signal ..., X511 carries out following calculating:

Zi＝Ci×Xi ------(1)；

(2) according to following formula (2) computation period cumulative signal Yi:

$\mathrm{Yi}=\underset{j=0}{\overset{7}{\mathrm{\&Sigma;}}}{Z}_{64j+i}---\left(2\right);$

(3) calculate subrane sample Si according to following formula (3):

$\mathrm{Si}=\underset{k=0}{\overset{63}{\mathrm{\&Sigma;}}}{Y}_k\&CenterDot;\mathrm{COS}(2i+1)(k-16)\mathrm{\&pi;}/64---\left(3\right);$

Subrane sample Si promptly is referred to as the subrane signal.

And in decoding equipment, when giving each passage with 32 subrane sample re-quantizations, synthetic filtering bank handles 32 audio samples Si in the operation step according to following formula (4)-(7):

(4) calculate Vi according to following formula (4) by 32 subrane samples of frequency shift Si:

$\mathrm{Vi}=\underset{k=0}{\overset{31}{\mathrm{\&Sigma;}}}{S}_k\&CenterDot;\mathrm{COS}(2i+1)(k+16)\mathrm{\&pi;}/64---\left(4\right);$

(5) ask for a series of Ui of 512 samples by the order of conversion Vi according to following formula (5):

U _64i+j＝V _128i+j

U _64i+32+j＝V _12i+96+j ------(5)；

(6) calculate Wi according to following formula (6) by multiply by impulse response with series of samples Ui:

Wi＝Ui×Di ------(6)；

(7) ask for output signal Sj according to following formula (7) by periodic accumulation:

$\mathrm{Sj}=\underset{i=0}{\overset{16}{\mathrm{\&Sigma;}}}{W}_{32i+j}---\left(7\right);$

In the aforementioned calculation process, data Vi and Ui are referred to as subrane synthetic filtering data, and output signal Sj is referred to as the PCM data.

The decoding that conventional decoding equipment 500 is carried out according to formula (4)-(7) is handled as shown in figure 45.

Decoding equipment 500 comprises a memory portion 510 and an operation part 520.Operation part 520 comprises that a subrane signal generation divides 521 and subrane composite parts 522, operation part is decoded into the coded signal of n passage input corresponding subrane signal and the subrane signal of each passage is carried out the operation of subrane synthetic filtering, thereby produces sound signal also with it output.Memory portion 510 is made up of a high speed access storage such as SRAM etc., and it comprises 511 to the 514 and storage areas 515 that are used for storing the subrane signal, storage area that are used for storing subrane synthetic filtering data.

Its operation of decoding equipment with this configuration is as follows.When coded signal at first was input to operation part 520, the subrane signal generation divided 521 coded signal is decoded into the subrane signal, and temporarily with the subrane signal storage in memory block 515.Then, subrane composite part 522 reads the subrane signal from memory block 515, and the subrane signal is carried out the operation of subrane synthetic filtering, thereby produces sound signal also with it output.

Utilization is stored in the subrane synthetic filtering data in the updated stored district, subrane synthetic filtering data division ground 511 to 514 that the subrane signal in the memory block 515 produced.Correspondingly, when carrying out the operation of subrane synthetic filtering, operation part 520 need read subrane synthetic filtering data from memory portion 510, and after operation is finished with subrane synthetic filtering data rewrite in memory portion 510.

Memory portion 510 must be can zero access storer.Under the situation of decoding equipment 500 corresponding to a plurality of passages, the situation of for example corresponding 4 passages, memory portion 510 must have 4 memory blocks 511 to 514, so that store the subrane synthetic filtering data that are used for 4 passages.

Be used for the high speed access storage of memory portion 510 such as SRAM etc., must have enough big capacity, for example, for the data of 4 passages of regenerating in real time, its capacity can store the subrane synthetic filtering data of 4 passages at least.Sort memory is generally all very expensive, thereby has improved the cost of audio frequency decoding device greatly.

Summary of the invention

According to an aspect of the present invention, provide a kind of, deciphered N by using subrane synthetic filtering data and subrane signal data to carry out the subrane synthetic operation _A(N _A＞1) audio frequency decoding device of the sound signal of individual passage.This decoding equipment comprises: a first memory part is used for storing M _A(M _A＜N _A) the subrane synthetic filtering data that are used for the subrane synthetic operation and the subrane signal data of individual passage; A second memory part is used for storing N at least _AThe subrane signal data and the N of individual passage _ASome data in the subrane synthetic filtering data of individual passage; An operation part is used for the voice data of received code and it is decoded into the subrane signal data, and utilizes the data that are stored in the first memory part to carry out the operation of subrane synthetic filtering, thus output M _AThe decoding audio data of individual passage, and require the position of conversion by the new subrane synthetic filtering data of subrane synthetic filtering operational computations and required next subrane synthetic filtering data; A tcp data segment is used for requirement according to operation part, partly transmits M in first memory part and second memory _AThe subrane synthetic filtering data of individual passage and subrane signal data.

In one embodiment of the invention, first memory partly comprises one first memory block and one second memory block, first memory block is used for storing the subrane synthetic filtering data and the subrane signal data of some passages, and second memory block is used for storing the subrane synthetic filtering data and the subrane signal data of another passage.Carry out i passage (i=1 is to N when utilization is stored in data in first memory block of first memory part _A, j=1 is to N _AAnd during the operation of the subrane synthetic filtering of i ≠ j), operation part will be stored in the data transmission of j passage in the second memory part in second memory block of first memory part, meanwhile, when utilization is stored in data in second memory block of first memory part when carrying out the subrane synthetic filtering operation of j passage, operation part will be stored in the k of second memory in partly, and (k=1 is to N _A, k ≠ i, the data transmission of the individual passage of and k ≠ j) in first memory block of first memory part, thereby carry out data transfer operation and the operation of subrane synthetic filtering concurrently.In another embodiment of the present invention, operation part comprises a continuous transfer instruction part, is used for coming a plurality of data transfer operations of order with an instruction when indication is carried out data transfer operation between first memory part and second memory part.

In another embodiment of the present invention, operation part comprises a virtual address distribution portion, and being used for provides virtual address after the actual address of the first memory memory block end points partly of storing subrane synthetic filtering data; The starting point of virtual address is distributed to actual address predetermined in the memory block; And sequentially other virtual address is distributed to actual address.

In yet another embodiment of the present invention, in first memory part, except that first memory block and second memory block, also comprise one the 3rd memory block, be used for storing and be not restricted to the data of using with the subrane synthetic operation.Operation part comprises a subrane signal hop, being used for port number when the input coding voice data is 3 or more for a long time, at least a portion is stored in subrane signal copy in the first memory part, that be used for special modality or is transferred to the special storage district of first memory part.

In yet another embodiment of the present invention, operation part comprises that a subrane signal adds/subtract part, be used for after having imported coding audio data and it is decoded into the subrane signal data, and subrane synthetic filtering data carried out adding/reducing of subrane signal concurrently from the transmission operation that second memory partly is transferred to the first memory part.

In yet another embodiment of the present invention, operation part comprises a staggered scanning part, be used for after finishing the subrane synthetic operation of last passage, with subrane synthetic filtering data are sampled to the decoding data of being deciphered by the subrane synthetic operation mutually concurrently from the transmission operation that first memory partly is transferred to second memory part, sample of each passage, and according to predetermined order sample is reset.

In yet another embodiment of the present invention, staggered scanning partly comprises a staggered scanning division part, is used for the staggered scanning division of operations is become the r step (r 〉=2).

In yet another embodiment of the present invention, staggered scanning partly comprises a staggered scanning storage area selection part, the port number that is used for deciphering according to the subrane synthetic operation is odd number or even number, partly selects a region of data storage that is used for the staggered scanning operation at first memory.

In yet another embodiment of the present invention, continuously transfer instruction comprises that partly a special data transmission finishes the detection indicating section, (q＞1, what the individual transmission of and 1≤p＜q) was operated finishes to be used to refer to p in q the data transfer operation of carrying out between first memory part and second memory part of detection.

In yet another embodiment of the present invention, tcp data segment comprises that the transmission of special data finishes the detected transmission part, be used for the instruction of partly sending according to continuous transfer instruction, finishing of p transmission operation in q the data transfer operation that detection is carried out between first memory part and second memory part, and the information of finishing of p the transmission operation that will detect sends to operation part.

In yet another embodiment of the present invention, operation part comprises that the transmission of special data finishes the test section, be used for the instruction of partly sending according to continuous transfer instruction, p that detects in q the data transfer operation of carrying out between first memory part and second memory part transmits finishing of operating.Finish after the data transfer operation that detects special section in the test section finishes in special data transmission, operation part is carried out s (r 〉=2, and 2≤s≤r) step operation in the r step staggered scanning operation.

In yet another embodiment of the present invention, transfer instruction partly comprises a PCM data transfer instruction part continuously, be used for when the port number of importing the decoding data of being deciphered by the subrane synthetic operation is t (t 〉=3), the decoding PCM data of at least one passage are transmitted in order between first memory part and second memory part.PCM data transfer instruction part partly is transferred to the second memory part from first memory with the PCM data that the subrane synthetic operation of last passage temporarily will be carried out the subrane synthetic operation mutually concurrently, and the PCM data that will transmit in the second memory part are transferred in the first memory part again.

In yet another embodiment of the present invention, transfer instruction partly comprises a PCM region of data storage selection part continuously, be used for partly selecting a storage area at first memory, port number according to the input decoding data of being deciphered by the subrane synthetic operation is odd number or even number, concurrently the PCM data partly is transferred to first memory operation partly from second memory mutually with the subrane synthetic operation of last passage.

In yet another embodiment of the present invention, operation part comprises a decoding division part, be used for decoding handled to be divided into occurring to output signal from the subrane signal this process taking place, maybe will decipher to handle to be divided into being synthesized to output signal from subrane this process takes place, thereby a plurality of audio output signal samples of each frame will be divided into y block.Here, a=b * c * y, in the formula: a represents the audio output signal sample number of each frame coding audio signal of each passage, the sample number that the partial wave hop count of b presentation code sound signal, c produce when representing to handle a block.

According to a further aspect of the invention, provide a kind of, deciphered N by using subrane synthetic filtering data and subrane signal data to carry out the subrane synthetic operation _A(N _A＞1) audio frequency decoding device of individual channel audio signal.This equipment comprises: a first memory part is used for storing subrane synthetic filtering data and the subrane signal data that at least one passage is used for the subrane synthetic operation; A second memory part is used for storing subrane signal data and N _AIndividual passage subrane synthetic filtering data; An operation part, be used for received code sound signal and coding audio signal is decoded into the subrane signal data, and utilize the data that are stored in the first memory part to carry out the operation of subrane synthetic filtering, thereby export the decoding audio data of a passage, and require the position of exchange by the new subrane synthetic filtering data and the required next subrane synthetic filtering data of subrane synthetic filtering operational computations; A tcp data segment is used for requirement according to operation part, and the exchange of passage ground is stored in first memory part and second memory subrane synthetic filtering data and the subrane signal data in partly one by one.

Therefore, foregoing invention has following advantage: (1) provides a kind of compressed code decoding device, be used for using a less relatively impact damper of storage volume to come the compressed code of decoding audio signal and with it output, the sound signal of output and timing signal and vision signal are synchronous; (2) provide a kind of audio frequency decoding device, this equipment is by using high speed access storage and the relatively low storer (for example DRAM) of access speed in memory portion, thereby have the ability of stronger processing frequency division coded system sound signal, thereby reduced the cost of the used storer of decoding coded signal.

By reading and understanding hereinafter in conjunction with the accompanying drawings detailed description, can clearly find out above-mentioned technological merit of the present invention and other advantage.

Description of drawings

Fig. 1 is the compressed code decoding device block diagram of explanation example 1 according to the present invention.

Fig. 2 is the data structure synoptic diagram that equipment input shown in Figure 1 is given in an explanation.

Fig. 3 is the compressed code decoding device block diagram of explanation example 2 according to the present invention.

Fig. 4 is the data structure synoptic diagram that equipment input shown in Figure 3 is given in an explanation.

Fig. 5 is the compressed code decoding device block diagram of explanation example 3 according to the present invention.

Fig. 6 is the data structure synoptic diagram that equipment input shown in Figure 5 is given in an explanation.

Fig. 7 is the buffer structure block diagram of an explanation in equipment shown in Figure 5.

Fig. 8 is the audio frequency decoding device block diagram of explanation example 4 according to the present invention.

Fig. 9 is the arrangement of explanation subrane synthetic filtering data and at the synoptic diagram (the 1st) of different time permutations.

Figure 10 is the arrangement of explanation subrane synthetic filtering data and at the synoptic diagram (the 2nd) of different time permutations.

Figure 11 is the arrangement of explanation subrane synthetic filtering data and at the synoptic diagram (the 3rd) of different time permutations.

Figure 12 is the figure of an explanation virtual address area and physical address block corresponding relation.

Figure 13 is the time diagram (the 1st) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 14 is the time diagram (the 2nd) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 15 is the time diagram (the 3rd) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 16 is the time diagram (the 4th) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 17 is the time diagram (the 5th) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 18 is the time diagram (the 6th) of a data transmission in the audio frequency decoding device of the example 4 according to the present invention.

Figure 19 is the audio frequency decoding device block diagram of explanation example 5 according to the present invention.

Figure 20 is the block diagram of the operation part of an explanation in equipment shown in Figure 19.

Figure 21 is the time diagram (the 1st) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 22 is the time diagram (the 2nd) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 23 is the time diagram (the 3rd) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 24 is the time diagram (the 4th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 25 is the time diagram (the 5th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 26 is the time diagram (the 6th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 27 is the time diagram (the 7th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 28 is the time diagram (the 8th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 29 is the time diagram (the 9th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 30 is the time diagram (the 10th) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 31 is the time diagram (the 11st) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 32 is the time diagram (the 12nd) of a data transmission in the audio frequency decoding device of the example 5 according to the present invention.

Figure 33 is the time diagram (the 1st) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 34 is the time diagram (the 2nd) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 35 is the time diagram (the 3rd) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 36 is the time diagram (the 4th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 37 is the time diagram (the 5th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 38 is the time diagram (the 6th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 39 is the time diagram (the 7th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 40 is the time diagram (the 8th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 41 is the time diagram (the 9th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 42 is the time diagram (the 10th) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 43 is the time diagram (the 11st) of a data transmission in the audio frequency decoding device of the example 6 according to the present invention.

Figure 44 is the example structure block diagram of a conventional compressed code decoding device of explanation.

Figure 45 is the example structure block diagram of a conventional audio frequency decoding device of explanation.

Embodiment

Hereinafter among Xu Shu the example 1-3, each example all provides a kind of compressed code decoding device of realizing the advantage (1) of compressed code decoding device, be used for by using the less relatively impact damper of memory capacity to come the compressed code of decoding audio signal and export decoded signal, the decoded signal of output and timing signal and vision signal are synchronous.

Each example among the example 4-6 all provides a kind of audio frequency decoding device of realizing the advantage (2) of audio frequency decoding device, this equipment is by using a high speed access storage and the storer (as DRAM) that access speed is relatively low in memory portion, thereby have the ability of stronger processing frequency division coded system sound signal, thereby reduced the cost of the used storer of decoding coded signal.

Example 1:

To narrate example 1 of the present invention with reference to the accompanying drawings hereinafter.Fig. 1 is the structured flowchart of explanation compressed code decoding device of example 1 according to the present invention.Fig. 2 is the example structure synoptic diagram that the combination compressed audio sign indicating number of compressed code decoding device input shown in Figure 1 is given in an explanation.

Can find out obviously that from Fig. 2 the compressed audio sign indicating number is combined into block.Block further is combined into signal frame, as frame A, B, C ...Each signal frame comprises a time mark T who is positioned at the signal frame reference position, is used to refer to the recovery time, is thereafter the compressed code of this signal frame.Recovery time when first block of compressed code comes the reproducing audio signal in the time mark T indication decoded frame.

The signal frame of a plurality of compressed codes forms signal packet, and compressed code is to transmit with the form of signal packet.Thereby the block of compressed code will be combined into signal frame and signal packet.

Signal frame shown in Figure 2 is input in the compressed code decoding device shown in Figure 1 by input terminal 1.Analysis circuit 5 reads the time mark T of each input signal frame in the reference position of signal frame, thereby obtains recovery time T0.Thereby analysis circuit 5 has just obtained the time of exporting the coded signal of first block in first signal frame.

Each block of compressed signal in the decoding scheme 2 decoded signal frames is so that output audio signal.Output buffer 3 temporarily stores the sound signal of at least two blocks.Output circuit 4 is externally exported the sound signal of reading from output buffer 3.

Whether testing circuit 6 detects all and externally exports by output circuit 4 corresponding to the sound signal of existing signal frame first block.Adding circuit 7 is according to the testing result of testing circuit 6 outputs, calculating begins time interval till export fully with the corresponding sound signal of first block from importing first block, the required time α of the first block sound signal of promptly regenerating, add the recovery time T0 that analysis circuit 5 obtains with time α then, thereby ask for lock in time (T0+ α).Timer 11 is given the clock timing and is exported current time R1.

The lock in time that synchronizing circuit 8 is asked for adding circuit 7 (T0+ α), the current time R1 with 11 timing of timer compared, if greater than current time R1, then determine that the regenerative operation of sound signal lags behind time schedule lock in time (T0+ α); If less than current time R1, then determine that the regenerative operation of sound signal is ahead of time schedule lock in time (T0+ α).

That is to say that timer 11 and synchronizing circuit 8 are used for compared with current time R1 lock in time (T0+ α), whether are ahead of current time R1 lock in time (T0+ α) so that determine.

(T0+ α＜R1), flip-flop circuit 9 is abandoned the next frame compressed code when synchronizing circuit 8 indications " hysteresis "; (T0+ α＞R1), flip-flop circuit 9 is not worked and when synchronizing circuit 8 indications " in advance ".When synchronizing circuit 8 outputs " in advance " (T0+ α＞R1), regeneration delays circuit 10 postpones the next down output time that passes through the sound signal piece of output circuit 4 outputs, and the time quantum of its delay is the time quantum that is ahead of current time R1 lock in time (T0+ α).This time adjustment is to be carried out by the output time of regeneration delays circuit control output buffer 3.

To narrate the operation of the compressed code decoding device of this example in more detail below.

At first, the signal packet by input terminal 1 incoming coded signal.Each signal packet comprises a plurality of signal frames, and each signal frame comprises a plurality of blocks.

Analysis circuit 5 extracts the header message of first signal frame and therefrom obtains the recovery time T0 of output corresponding to the corresponding sound signal of first block.And decoding scheme 2 receives the block of the first signal frame compressed code from analysis circuit 5, and deciphers compressed code block by block, thus output audio signal.

The sound signal of first block of decoding scheme 2 decodings temporarily is stored in the output buffer 3.When sound signals all in the individual signals piece all has been ready to, externally export the sound signal of this block by output circuit 4.When output circuit 4 begins to export the sound signal of first block, begin the sound signal of second block is write in the output buffer 3 successively.After the sound signal of first block is exported fully, export the sound signal of secondary signal piece then.

Testing result according to testing circuit 6 outputs, adding circuit 7 calculates regeneration and exports the required time α of the first block sound signal, output time α with first block adds the time T 0 that analysis circuit 5 obtains then, thereby tries to achieve lock in time (T0+ α).Synchronizing circuit 8 is asked for lock in time (T0+ α) and poor between the time R1 of timer 11, and judges that whether poor (T0+ the α)-R1 that is tried to achieve is greater than predetermined value TH.

When (T0+ the α)-R1 that is tried to achieve differs from greater than predetermined value TH, thereby satisfy following formula (8):

(T0+α)-R1＞TH ------(8)

Then determine to be ahead of current time R1 lock in time (T0+ α).

When (T0+ the α)-R1 that is tried to achieve differs from less than predetermined value TH, thereby satisfy following formula (9):

(T0+α)-R1＜TH ------(9)

Then determine to lag behind current time R1 lock in time (T0+ α).

When definite result of synchronizing circuit 8 satisfied above-mentioned formula (8), when analysis circuit 5 inputed to decoding scheme 2, flip-flop circuit 9 was abandoned the second frame compressed code in the second frame compressed code.When satisfying above-mentioned formula (9), flip-flop circuit 9 is not worked.

And when definite result of synchronizing circuit 8 satisfied above-mentioned formula (9), regeneration delays circuit 10 was gone up the sound signal of the 3rd block of output first signal frame in the time that has postponed some time quantums (i.e. the difference of (T0+ α)-R1).When satisfying above-mentioned formula (8), regeneration delays circuit 10 is not worked.

Above-mentioned control operation can be according to the recovery time output audio signal that offers signal frame.Provide this example described compressed code decoding device structure by the decoding equipment of giving the decoding compression video code, can pass through the mutual synchronously reproducing Voice ﹠ Video of corresponding apparatus sign indicating number.In this case, provide the recovery time data to video codes, thereby make video codes and audio code synchronised, meanwhile, video decoding equipment uses timer or the shared timer 11 identical with timer 11 types.

According to this example, only require the capacity and two identical getting final product of block capacity of output buffer 3, thereby can reduce the data volume that will store.

In above-mentioned example, whether testing circuit 6 detect corresponding to the first block sound signal and export the outside.This block is not limited only to first block, also can one any other block.Yet note: the output time α of the block that adding circuit 7 is added up must be number corresponding with block.

In detail, when the block quantity of finishing output of testing circuit 6 detections was n, the value of adding circuit 7 added α was:

α=(the output deadline of a sound signal piece) * n

And the recovery time T0 that analysis circuit 5 is read is the recovery time of the next signal frame of decoded current demand signal frame back.Therefore, even, also can realize similar effect when obtaining when being used for the adding circuit 7 time α with block number corresponding block that add up and being negative value.Therefore, when the block quantity of finishing output of testing circuit 6 detections was n, the value of the α that adding circuit 7 adds up was:

α=-(the output deadline of a sound signal piece) * (6-n)

Example 2

Hereinafter, the compressed code decoding device of the example 2 according to the present invention will be narrated in conjunction with the accompanying drawings.Fig. 3 is the compressed code decoding device structured flowchart of an illustrated example 2.Fig. 4 is the example structure synoptic diagram that an explanation inputs to the combination compressed audio sign indicating number of compressed code decoding device shown in Figure 3.

As shown in Figure 4, in the equipment of this example, suppose compressed audio sign indicating number and compression video code parallel transmission, and signal frame A, the B of each compressed audio sign indicating number, C ... recovery time and compression video code signal frame a, b, c ... the recovery time synchronised.

Compare with the equipment of Fig. 1, compressed code decoding device shown in Figure 3 also provides a video analysis circuit 12, video coding circuit 13, video output circuit 14, a video refresh memory time memory circuit 15 and a video input terminal 16 extraly.

The function of elements such as decoding scheme 2, output buffer 3, output circuit 4, analysis circuit 5, testing circuit 6, adding circuit 7, flip-flop circuit 9, regeneration delays circuit 10 and timer 11 is identical with the element function in Fig. 1 equipment.

Video coding circuit 13 is deciphered the video codes of compression, and externally exports decoding video signals by video output circuit 14.

When the compressed audio sign indicating number that is combined into signal frame and compression video code inputed to video analysis circuit 12, video analysis circuit 12 obtained the recovery time V0 of regeneration vision signal from the reference position of each signal frame.Video refresh memory time memory circuit 15 calculates and stores video refresh memory time V0 and from the mistiming (V0-R1) between the current time R1 of timer 11.Video coding circuit 13 receives compression video code and it is deciphered.Video output circuit 14 is externally exported the vision signal of decoding.

In this structure, video analysis circuit 12 obtains the video refresh memory time V0 of regenerated signal frame.Video refresh memory time memory circuit 15 obtains and stores video refresh memory time V0 and from the mistiming (V0-R2) between the current time R2 of timer 11.

When testing circuit 6 detects first sound signal piece when having exported fully, synchronizing circuit 8 the time (V0-R2) in the video refresh memory time memory circuit 15 of will being stored in adds the time R1 from timer 11, thereby ask for time T V (=V0-R2+R1), and further ask for difference (T0+ α)-TV between the lock in time (T0+ α) that time T V and adding circuit 7 obtain.

According to above-mentioned difference (T0+ α)-TV, the recovery time of synchronizing circuit 8 definite sound signals is in advance or lags behind the arrangement of time of design.

When the value of (T0+ α)-TV during, promptly satisfy greater than predetermined value TH

(T0+α)-TV＞TH ……(10)

Then determine to be ahead of current time R1 lock in time (T0+ α).

When the value of (T0+ α)-TV during, promptly satisfy less than predetermined value TH

(T0+ α)-TV＜TH ... (11) then determine to lag behind current time R1 lock in time (T0+ α).

Attention: in these 13 decoding required times of compression video code of hypothesis video coding circuit is 0, and video refresh memory time V0 represents the time that vision signal is exported from video output circuit 14.

In this example, compression video code is transfused to video input terminal 16.Yet, when comprising that other compressed code such as benchmark recovery time replaces compression video code and imports with the compressed audio sign indicating number, the present invention can be used for regenerating compressed audio sign indicating number and other compressed code of phase mutually synchronization.Unique difference is to use the benchmark recovery time to replace video refresh memory time V0.

Example 3

Hereinafter, the compressed code decoding device of the example 3 according to the present invention will be narrated in conjunction with the accompanying drawings.Fig. 5 is the compressed code decoding device structured flowchart of an illustrated example 3.Fig. 6 is the example structure synoptic diagram that an explanation inputs to the combination compressed audio sign indicating number of compressed code decoding device shown in Figure 5.

As shown in Figure 6, in this example, suppose in the signal frame of transmission continuously, the blank time section X of signal frame to occur not having with a certain predetermined interval.The Equipment Inspection of this example goes out blank time period X and next signal frame C that regenerates after wait and the corresponding time span of blank time section X.For example, when transmitting compressed audio sign indicating number and compression video code concurrently, because the information that compression video code comprises more than compressed audio sign indicating number (so the length of video codes is longer), blank time section X therefore will occur between two compressed audio sign indicating numbers.Therefore must be after wait and the corresponding time x of blank time section X, regeneration next signal frame C.

Compare with the equipment of Fig. 1, compressed code decoding device shown in Figure 5 provides a time interval extraly and circuit 17, block testing circuit 18, block adding circuit 19, time detection circuit 20 and one is set according to time interval regeneration delays circuit 21.

The function of elements such as decoding scheme 2, output buffer 3 and output circuit 4 is identical with the function of element shown in Figure 1.

The zero-time Ts in the time interval and concluding time Tf (corresponding to beginning and the end of time period x) externally offer or are arranged on the time interval and be provided with in the circuit 17.Whether block testing circuit 18 detects the sound signal piece and exports fully from output circuit 4.

Analysis circuit 5 obtains the recovery time T0 of first block of first signal frame from first signal frame, and recovery time T0 is offered block adding circuit 19.Whenever block testing circuit 18 detects some sound signal pieces when having exported fully, block adding circuit 19 obtains the required time β of this block of regeneration, and add time β with recovery time T0, add the required time α of regeneration first block, thereby ask for the time (T0+ alpha+beta).

Time detection circuit 20 detects and is arranged on zero-time Ts that the time interval is provided with the time interval in the circuit 17 whether between the time (T0+2 alpha+beta) of time (T0+ alpha+beta) that block adding circuit 19 obtains and the time α gained by first block that adds up again.When 20 detection times of time testing circuit during at interval zero-time Ts, will descend next block to be deferred to the concluding time Tf in the time interval by the output time of output circuit 4 according to time interval regeneration delays circuit 21.

For example, as recovery time T0 from first block of first signal frame of signal packet, when having experienced the time span between time (T0+ alpha+beta) and time (T0+2 alpha+beta), the output time of first block in first signal frame of next signal bag is deferred to the concluding time Tf in the time interval.According to the method, time period x is inserted in the sound signal of regeneration.

Fig. 7 is the structured flowchart of the output buffer 3 of key diagram 5 apparatus shown.Output buffer 3 comprises a first memory circuit 22, second memory circuit 23, status circuit 24, a write control circuit 25 and a state control circuit 26.

The decode results of first memory circuit 22 and second memory circuit 23 storage decoding schemes 2.The sound signal of the outside output of status circuit 24 indications is from 22 outputs of first memory circuit or from 23 outputs of second memory circuit.Write control circuit 25 executivecontrol functions, some status circuits that the sound signal that decoding scheme 2 is deciphered is written in first memory circuit 22 and the second memory circuit 23 do not have in the circuit of indication.

When block testing circuit 18 detects some sound signal pieces and exports fully, state control circuit 26 switches to another memory circuitry from a memory circuitry storage writes the memory circuitry of decode results from the memory circuitry of the sound signal of output circuit 4 and write control circuit 25 among.When block testing circuit 18 detects some sound signal pieces and has exported fully, in the read operation and write operation of switching first and second memory circuitries, write control circuit 25 and state control circuit 26 are controlled the write operation that the sound signal piece is written to the memory circuitry that state control circuit 26 do not indicate.

In this structure, after first block is decoded, write control circuit 25 is written to the decode results of first block in the first memory circuit 22 immediately, and after the secondary signal piece was decoded, the decode results with the secondary signal piece was written in the second memory circuit 23 immediately.Then, in case all decode results all have been written in the first memory circuit 22, output circuit 4 begins output signal externally.Like this, continue to carry out above-mentioned write operation and export the 3rd and the decode results of follow-up signal piece.

In fact, above-mentioned each example is all by using a plurality of discrete ICs, DSP (digital signal processor) or CPU to realize.

Above-mentioned example 1 has following effect to example 3 described compressed code decoding devices:

Do not need to use the mass storage that is used for storing a plurality of signal frames or signal packet, thereby can reduce the circuit scale of AV regeneration equipment, and can decipher and the compressed audio sign indicating number of regeneration and timing signal or video image synchronised.

Example 4

The audio frequency decoding device of example 4 receives the multi-channel coding signal that is used for the MPEG2 second layer (the layer 2 of MPEG) (port number N in this example _A=4), and with coded signal be decoded into sound signal.In ISO/IEC 11172-3:1993 and 13818-3:1996, the MPEG2 second layer is described in detail.

Fig. 8 be one according to the present invention the block diagram of the audio frequency decoding device 100 of example 4.Audio frequency decoding device 100 comprises a first memory part 110, second memory part 120, an operation part 130 and a Data Transmission Control Unit 140.First memory part 110 comprise one can zero access SRAM and can store M _A(M _A＜N _A) the subrane synthetic filtering data and the subrane signal of individual passage.In this example, M _A=2.First memory part 110 is divided into a memory block 111 (first memory block) and a memory block 112 (second memory block).The signal of interest processing section of audio frequency decoding device 100 comprises the monolithic multimedia processor, and this processor comprises an internal memory.When internal memory is mainly used in pictorial data when handling, first memory part 110 is memory blocks of internal memory, is used for audio signal.

The storage of first memory block 111 be used for X passage voice data (X be one from 1 to N _ABetween integer) and comprise two memory block 111A and 111B.111A storage in memory block is used for the subrane synthetic filtering data (corresponding to handling " V " that produces according to mpeg standard by being used for the synthetic matrix of subrane) of X passage, and 111B storage in memory block is used for X passage subrane signal data.The storage of second memory block 112 be used for Y passage voice data (wherein Y be one from 1 to N _ABetween integer, but be not equal to X) and comprise two memory block 112A and 112B.112A storage in memory block is used for the subrane synthetic filtering data of Y passage, and 112B storage in memory block is used for Y passage subrane signal data.In the narration below, N _A=4.

Second memory part 120 is a storer, and it comprises a both cheap and DRAM capacious who is connected on the processor outward.Second memory part 120 comprises that memory block 121A is to 124A, 123B and 124B.The subrane synthetic filtering data of each the storage area stores passage of memory block 121A in the 124A.For example, the subrane synthetic filtering data storage that is used for first passage is at memory block 121A, the subrane synthetic filtering data storage that is used for second passage is at memory block 122A, the subrane synthetic filtering data storage that is used for the 3rd passage is at memory block 123A, and the subrane synthetic filtering data storage that is used for the 4th passage is at memory block 124A.

The subrane signal data that is used for the 3rd passage is stored in memory block 123B, and the subrane signal data that is used for the 4th passage is stored in memory block 124B.

Memory block 111A in first memory part 110 and 112A and the memory block 121A in second memory part 120 are big or small identical to 124A's.In detail, each memory block in these memory blocks can comprise address 1 to 2048 (hexadecimal 0x800).

Memory block 111B in first memory part 110 and 112B and memory block 123B in second memory part 120 and 124B's is big or small identical.In detail, each memory block in these memory blocks can comprise address 1 to 768 (hexadecimal 0x300).

Operation part 130 comprises that a subrane signal generation divides 131, subrane composite parts 132, a continuous transfer instruction part 133 and a virtual address distribution portion 134.

The subrane signal generation divides 131 will be decoded into the subrane signal from the coded signal of peripheral hardware input.Subrane composite part 132 utilizes subrane synthetic filtering data and subrane signal to carry out the operation of subrane synthetic filtering in passage ground one by one, thereby produces sound signal.Transfer instruction part 133 is specified special memory block in each memory portion of first memory part 110 and second memory part 120 continuously, and director data transmission control unit (TCU) 140 is carried out one or more data transmission.

There is a virtual address in virtual address distribution portion 134 hypothesis after the actual address of each memory block 111A and 112A end points, and the starting point of virtual address are distributed to the actual address of each memory block 111A and 112A reference position.Virtual address distribution portion 134 is sequentially distributed to virtual address the actual address of back by this way.

The instruction that Data Transmission Control Unit 140 sends according to continuous transfer instruction part 133 is transferred to data second memory part 120 or is transferred to the first memory part 110 from second memory part 120 from first memory part 110.

Operation part 130 orders are carried out data transmission between the second memory part 120 of some storeies among memory block 111 and 112 and external processor.During data transmission, operation part utilizes another memory block in memory block 111 and 112 to carry out the subrane synthetic filtering operation of a passage.

Therefore, the transmission of the generation of sound signal and data is parallel processings.For example, when transmission data between first memory block 111 of first memory part 110 and second memory 120, operation part 130 is utilized the data executable operations in second memory block 112 of first memory part 110.When carrying out data transmission between second memory block 112 and second memory part 120, operation part 130 is utilized the data executable operations in first memory block 111.

As the form of narrating in ISO/IEC11172-3:1993 and 13818-3:1996, it is 32 that the subrane signal generation divides the subrane number of signals of 131 each passage that once produces.The sample number of the subrane sample of signal of a signal frame is 1152 in every passage.It is 32 * N that the subrane signal generation divides 131 each subrane sample of signal of producing for each passage to count Sn, and N is the integer from 1 to 36 (comprising 36).

And it is 32 in every passage that subrane composite part 132 is carried out the required minimum subrane sample of signal number of subrane synthetic filtering operation.At first, carry out matrix and handle,, utilize the subrane signal of 32 samples and cosine function to upgrade 1/16 of continuous subrane synthetic filtering data according to above-mentioned formula (4).Then, carry out multiplication-sum operation, utilize the subrane synthetic filtering data Ui and the window mouth coefficient Di that upgrade to produce 32 sample audio signal according to above-mentioned formula (6) and (7).Whenever obtain 32 sample audio signal, carry out multiplication-sum operation 16 times.At this moment, from up-to-date sample, a sample connects a sample order ground and gives the subrane synthetic filtering data assignment that is used to operate.In detail, at first give the data assignment corresponding to the subrane synthetic filtering data 1/16 of latest update, give 1/16 data assignment of the subrane synthetic filtering data upgraded corresponding to nearest previous operation then, the rest may be inferred.Therefore, execution graph 9 is to processing operation shown in Figure 11.

In this example, for the purpose of simplifying the description, suppose that the value of N is set to 6.In detail, to divide the sample number of the 131 subrane signals that once produce be 192 (32 * 6) to the subrane signal generation.Subrane composite part 132 utilizes the subrane signal, carries out 6 subrane synthetic filtering operations at each passage, thereby produces the sound signal of 192 (32 * 6) samples.The subrane synthetic filtering data that the operation of subrane synthetic filtering is upgraded account for 6/16 of whole subrane synthetic filtering data.

Fig. 9 shows the arrangement of the subrane synthetic filtering data that produce at each passage and the variation of arranging to Figure 11 in the whole time period.In Figure 11, a plurality of rectangle regions 200 are arranged at Fig. 9, each rectangle region is represented whole subrane synthetic filtering data.Subrane synthetic filtering data 200 are divided into 16 data fields 201 to 216.Reference number 222-1 below 200 districts is illustrated in variation order in the whole time period to 222-16.The time interval between contiguous two rectangle regions 200 is corresponding to carrying out a required time of subrane synthetic filtering data.The arrangement of data 222-1 is the oldest in the whole time period, and the arrangement of data 222-16 is up-to-date.In this example, Fig. 9 shows arrangement and the position of data 222-1 to the pointer P2 of 222-6, and Figure 10 shows arrangement and the position of data 222-7 to the pointer P2 of 222-12, and Figure 11 shows arrangement and the position of data 222-13 to the pointer P2 of 222-16.

Arrow on each rectangular block 200 right side is represented pointer P2, and indication is upgraded in the data in this district.As mentioned above, at every turn the sample number of the subrane signal that each passage is produced is 192 (32 * 6), corresponding to carrying out 6 subrane synthetic filterings operations.The subrane synthetic filtering data that tentation data 222-1 is upgraded for the first subrane synthetic operation of carrying out immediately after the subrane signal takes place, during changing from data 222-6 to data 222-7 and during changing to data 222-13 from data 222-12, the subrane signal generation divides the subrane signal of 192 samples of 131 generations.During this period, Data Transmission Control Unit 140 control datas transmit between first memory part 110 and second memory part 120.When subrane synthetic filtering data are in state shown in the rectangular block 222-1, the data in the data field 201 have just been upgraded.When subrane synthetic filtering data are in state shown in the rectangular block 222-2, the data in the data field 202 have just been upgraded.The rest may be inferred, when subrane synthetic filtering data are in rectangular block 222-3 to the position shown in the 222-16, just upgraded the data in the data field 203 to 216 respectively, and then upgrade the data among the 222-1 of data field.

Figure 12 is that an expression is distributed to the virtual address of first memory part 110 and the mapping table between the actual address virtually by virtual address distribution portion 134 shown in Figure 8.As can be seen from Figure 12, the address that is included in memory block 111 and 112 is respectively 0 to 3071 (sexadecimal is that 0x000 is to 0xbff).In virtual address is 0x000 during to 0xbff, and virtual address is identical with the value of actual address.Yet virtual address 0xc00 represents respectively that to 0xfff actual address 0x000 is to 0x7ff.

For example, the memory block 111B of first memory block 111 is dispensed on address 0x100 to 0x3ff, and memory block 111A is dispensed on address 0x400 to 0xbff.The rest may be inferred, and the memory block 112B of second memory block 112 is dispensed on address 0x100 to 0x3ff, and memory block 112A is dispensed on address 0x400 to 0xbff.Because the virtual address 0xc00 in each memory block 111 and 112 all is assigned to actual address 0x400, so virtual address 0xc00 represents the leading address of each memory block 111A and 112A.And the virtual address of back is represented actual address in an identical manner.

Figure 13 is the time diagram of data transmission to Figure 18.Figure 13 shows transmission diagram 1 to 6 respectively to Figure 18.Transmission diagram 1 shown in Figure 13 shows the data transmission before subrane is synthetic, transmission diagram 2 shown in Figure 14 shows the data transmission of subrane between synthesis phase at first passage, transmission diagram 3 shown in Figure 15 shows the data transmission of subrane between synthesis phase at second passage, and transmission diagram 4 shown in Figure 16 shows the data transmission of subrane between synthesis phase at the 3rd passage.The transmission diagram 5 of Figure 17 shows in the data transmission of the 4th passage subrane between synthesis phase, and the transmission diagram 6 in Figure 18 shows the data transmission after subrane is synthetic.

The operation of audio frequency decoding device 100 with said structure is as follows:

At first, when importing hyperchannel (4 passages) bit stream (sound signal that comprises coding) of MPEG2, operation part 130 flows to the subrane signal generation with bit stream and divides 131, and the subrane signal generation divides 131 bit stream is decoded into the subrane signal of 4 passages.Then, the subrane signal of the individual passage of i (i=1 in this example) is written among the 111B of memory block, the subrane signal of the individual passage of j (j=2 in this example) is written among the 112B of memory block.The subrane signal of the individual passage of k (k=3 in this example) is written in the part memory block of memory block 111A, and the subrane signal of the individual passage of l (l=4 in this example) is written in the part memory block of memory block 112A.

Then, transfer instruction part 133 is according to the requirement of operation part 130 continuously, and order data transmission control unit (TCU) 140 transmits data in the following manner.As transmit (Figure 13) shown in Figure 1, the subrane signal data that is stored in the third channel among the memory block 111A of first memory part 110 is transferred among the memory block 123B of second memory part 120, will be stored in the subrane synthetic filtering data transmission of the first passage among the memory block 121A of second memory part 120 in the memory block 111A of first memory part 110.

Therefore, Data Transmission Control Unit 140 is transferred to the subrane signal data of third channel the 123B of memory block from memory block 111A, and afterwards, existing side by side soon, the subrane synthetic filtering data of first passage are transferred to the 111A of memory block from memory block 121A.When finishing data transmission, Data Transmission Control Unit 140 notifying operation parts 130 have been finished the transmission of data.

Operation part 130 asks continuous transfer instruction part 133 to transmit data in following mode after determining to have finished data transmission.

As transmit (Figure 14) shown in Figure 2, the subrane signal data that is stored in the four-way among the memory block 112A of first memory part 110 is transferred among the memory block 124B of second memory part 120, will be stored in the subrane synthetic filtering data transmission of the second channel among the memory block 122A of second memory part 120 in the memory block 112A of first memory part 110.Operation part 130 is operated the data of utilizing concurrently in first memory block 111 with above-mentioned transmission, starts the subrane synthetic filtering operation of first passages by subrane composite part 132.

At this moment, subrane synthetic filtering data in the data field 201 of data 222-1 (Fig. 9) have been upgraded.In other words, upgraded address 0x680 in first memory block 111 to the data of 0x6ff.Afterwards, ask for the sound signal of 32 samples by carrying out 16 multiplication-sum operation.For the assignment consistance in proper order of assurance coefficient D, from the sample of final updating, the subrane synthetic filtering data assignment that is used to operate is given on sample order ground one by one.In detail, at first giving the subrane synthetic filtering data corresponding to final updating is 1/16 assignment of the data in the data field 201, next then give corresponding near on subrane synthetic filtering data of upgrading before upgrading operation be 1/16 sample assignment of the data in the data field 216, the rest may be inferred, give in the data field 215,214 ..., until 203 and 202 in the data assignment.

After the subrane synthetic filtering data assignment in giving data field 207 (0xb80 is to 0xbff), to data field 206 and after during the subrane synthetic filtering data assignment of (0x400-), for the address is set, need carry out the circulation address process to actual address.The circulation address process is operated by following logical multiply and is carried out.At this, the address before the conversion is A, and the address B after the conversion is:

B＝{(A-0x400)×(0x7ff)}+0x400

In the formula * presentation logic takes advantage of operation.

Yet in this example, operation part 130 includes virtual address distribution portion 134, thereby does not need to carry out the circulation address process that comprises logical multiply.Utilizing virtual address 0xc00 can distribute actual address 0x400 to 0xfff is subrane synthetic filtering data in data field 206 to 201,216 and 215 to 0x7ff.

Shown in rectangle region 222-2 (Fig. 9), in follow-up subrane synthetic filtering operation, also upgrade the data in the data field 202, and utilize the subrane synthetic filtering data in data field 201,216 to 207 and 206 to 202 to carry out multiplication-sum operation, thus the sound signal of obtaining.

And, shown in rectangle region 222-3 (Fig. 9), in next step subrane synthetic filtering operation, also upgrade the data in the data field 203, and utilize the subrane synthetic filtering data in data field 202,201,216 to 207 and 206 to 203 to carry out multiplication-sum operation, thereby the sound signal of obtaining.

Shown in rectangle region 222-6 (Fig. 9), in the 6th step subrane synthetic filtering operation before transmitting audio signal, also upgrade the data in the data field 206, and utilize the subrane synthetic filtering data in data field 205 to 201,216 to 207 and 206 to carry out multiplication-sum operation, thereby the sound signal of obtaining.In sum, in carrying out 6 subrane synthetic filtering operations, do not need to carry out multiplication-sum operation that the circulation address process just can be carried out usage factor D.

When operation part 130 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 140 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 140 notifying operation parts 130 had been finished data transmission.When the data transmission of having finished first passage and the operation of subrane synthetic filtering, transfer instruction part 133 is according to the requirement of operation part 130 continuously, and order data transmission control unit (TCU) 140 transmits data in the following manner.

As transmit (Figure 15) shown in Figure 3,133 orders of transfer instruction part will be stored in first passage subrane synthetic filtering data transmission among the memory block 111A of first memory part 110 in the memory block 121A of second memory part 120 continuously.Continuously the 133 further orders of transfer instruction part will be stored in third channel subrane synthetic filtering data transmission among the memory block 123A of second memory part 120 in the memory block 111A of first memory part 110.Transfer instruction part 133 also orders the third channel subrane signal data among the memory block 123B that will be stored in second memory part 120 to be transferred among the memory block 111B of first memory part 110 continuously.Subrane composite part 132 utilizes the data of second memory block 112 to start the subrane synthetic filtering operation of second channel.

The transmission of the subrane synthetic filtering data of first passage and third channel is carried out between rectangle region 222-6 (Fig. 9) and rectangle region 222-7 (Figure 10).The following execution of subrane synthetic filtering data transmission of first passage.Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.By this operation, the address assignment order of subrane synthetic filtering data that is used for the data transmission after the first passage is identical with former transmission.

In this stage, utilize second memory block 112, according to operating identical method, carry out the subrane synthetic filtering operation of second channel with carrying out first passage subrane synthetic filtering.When subrane composite part 132 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 140 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 140 notifying operation parts 130 had been finished data transmission.

When the data transmission of finishing second channel and the operation of subrane synthetic filtering, operation part 130 requires continuous transfer instruction part 133 according to following method transmission data.

Shown in transmission diagram 4 (Figure 16), continuously 133 orders of transfer instruction part will be stored in second channel subrane synthetic filtering data transmission among the memory block 112A of first memory part 110 in the memory block 122A of second memory part 120.Continuously the 133 further orders of transfer instruction part will be stored in four-way subrane synthetic filtering data transmission among the memory block 124A of second memory part 120 in the memory block 112A of first memory part 110.Then, continuously 133 orders of transfer instruction part will be stored in the memory block 112B that the subrane signal data that is used for four-way among the memory block 124B of second memory part 120 is transferred to first memory part 110.Subrane composite part 132 utilizes the data of first memory block 111 and the subrane synthetic filtering operation that above-mentioned transmission operation starts third channel concurrently.

The transmission of the subrane synthetic filtering data of second channel and four-way is according to the mode identical with the transmission between first passage and third channel, carries out between rectangle region 222-6 (Fig. 9) and rectangle region 222-7 (Figure 10).Rectangle region 200 is divided into A1 and A2 two parts (Fig. 9), and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.

When the data transmission of finishing third channel and the operation of subrane synthetic filtering, operation part 130 requires continuous transfer instruction part 133 according to following method transmission data.

Shown in transmission diagram 5 (Figure 17), will be stored in the subrane synthetic filtering data transmission that is used for third channel among the memory block 111A of first memory part 110 to the memory block 123A of second memory part 120.Subrane composite part 132 utilizes the data of second memory block 112, starts the subrane synthetic filtering operation of four-way.

When the data transmission of finishing four-way and the operation of subrane synthetic filtering, operation part 130 requires continuous transfer instruction part 133 according to following method transmission data.

Shown in transmission diagram 6 (Figure 18), continuously 133 orders of transfer instruction part will be stored in four-way subrane synthetic filtering data transmission among the memory block 112A of first memory part 110 in the memory block 124A of second memory part 120.

Then, at next step, the subrane signal generation divides 131 to produce 192 sample subrane signals that are used for each passage.Afterwards, carry out subrane synthetic operation and data transfer operation concurrently.Be used for any passage subrane synthetic filtering data are carried out between rectangle region 222-12 (Figure 10) and rectangle region 222-13 (Figure 11) from the transmission operation that first memory part 110 is transferred to second memory part 120.Rectangle region 200 is divided into B1 and B2 two parts (Figure 10), and B1 comprises data field 212 to 203, and B2 comprises data field 202,201 and 216 to 213, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.

From top narration as can be seen, in the audio frequency decoding device 100 of this example, used DRAM that a cost is lower than SRAM, be used for storing the data of all passages in the corresponding sub-second memory part 120 as second memory part 120.Only just use when needed Data Transmission Control Unit 140 with the data transmission in the second memory part 120 in first memory part 110.Therefore, need not to increase the capacity of processor memory and the multi-channel audio signal of can regenerating at high speed.Owing to do not need external SRAM, therefore can reduce the cost of entire equipment.

In the audio frequency decoding device 100 of this example, can operate and data transfer operation by executed in parallel subrane synthetic filtering, therefore can shorten the required time of operational processes by the required time of data transmission.Owing to can transmit the data volume of operating by single transmission operation, thereby in the subrane synthetic operation that has used the virtual address distribution portion, not need to carry out circulation address process and subrane synthetic filtering data shift and handle corresponding to a plurality of transmission.And, because data transmission can be carried out repeatedly, therefore can and produce the required time transmission of arranging data at an easy rate of sound signal according to data transmission.Therefore, can reduce the loss of time that data transmission causes, thereby can shorten the processing time.

In this example, in order to simplify narration, supposing to be used for the port number of input signal is 4.Yet the quantity of passage is not limited to 4.

In this example, any possible method can be used for data transmission before the subrane of first passage is synthetic concurrently or with the subrane synthetic filtering data transmission processing of execution concurrently mutually of last passage.For example, after the subrane synthetic filtering data of last passage of transmission, can produce the subrane signal by input signal.Like this, can reduce the loss that data transmission causes.

Example 5

The audio frequency decoding device of example 5 receives the coded signal of the hyperchannel (5 passages) that is used for the MPEG2 second layer, and coded signal is decoded into sound signal.In ISO/IEC 11172-3:1993 and 13818-3:1996, the second layer of MPEG2 is described in detail.

Figure 19 be one according to the present invention the block diagram of the audio frequency decoding device 300 of example 5.Audio frequency decoding device 300 comprises a first memory part 310, second memory part 320, an operation part 330 and a Data Transmission Control Unit 340.First memory part 310 comprise one can zero access SRAM and be divided into a memory block 311 (first memory block), a memory block 312 (second memory block) and a memory block 313 (the 3rd memory block).The signal of interest processing section of audio frequency decoding device 300 comprises the monolithic multimedia processor, and this processor comprises an internal memory.When internal memory is mainly used in pictorial data when handling, first memory part 310 is memory blocks of internal memory, and deposit is used for audio signal.

The storage of first memory block 311 be used for X passage voice data (wherein X be one from 1 to N _ABetween integer) and comprise two memory block 311A and 311B.Memory block 311B mainly stores the subrane signal that is used for X passage, and memory block 311A mainly stores the subrane synthetic filtering data that are used for X passage.

The storage of second memory block 312 be used for Y passage coding audio data (wherein Y be one from 1 to N _ABetween integer, but be not equal to X) and comprise two memory block 312A and 312B.Memory block 312C is among the 312A of memory block.

Memory block 312B mainly stores the subrane signal that is used for Y passage, and memory block 312A mainly stores the subrane synthetic filtering data that are used for Y passage.Memory block 312C be operation part 330 provisionally (being to have produced from the subrane signal to begin to finish this section period at least) to first subrane of some passages is synthetic store some memory blocks by the synthetic subrane signal that is used for another passage that produces of the 3rd subrane.

Memory block 313 is memory blocks that are used for storing such as the input signal header message.There is a memory block 313C memory block 313, and operation part 330 can be therein provisionally (being to have produced from the subrane signal to begin to finish this section period to first subrane of some passages is synthetic at least) stores the subrane signal of other passage.In the narration hereinafter, N _A=5.

Second memory part 320 is one and comprises the outer storer that is connected on the high capacity DRAM on the processor.Second memory part 320 be divided into memory block 321A to 325A, 323B to 325B, 321C is to 323C and 326.

At memory block 321A in 325A, the subrane synthetic filtering data of a passage of each storage area stores.For example, the subrane synthetic filtering data storage that is used for first passage is at memory block 321A, the subrane synthetic filtering data storage that is used for second channel is at memory block 322A, the subrane synthetic filtering data storage that is used for third channel is at memory block 323A, the subrane synthetic filtering data storage that is used for four-way is at memory block 324A, and the subrane synthetic filtering data storage that is used for the five-way road is at memory block 325A.

The part subrane signal data that is used for third channel is stored in memory block 323B, and the subrane signal data that is used for four-way is stored in memory block 324B, and the subrane signal data that is used for the five-way road is stored in memory block 325B.

321C storage in memory block is used for the PCM data of first passage, and 322C storage in memory block is used for the subrane signal of second channel, and 323C storage in memory block is used for the subrane signal of third channel.

Memory block 326 is used as the audio signal buffer device, and is used for storing the staggered scanning data, in the staggered scanning data, first to the PCM data in five-way road according to sound signal similar data format permutation.

Memory block 311A in first memory part 310 and 312A and the memory block 321A in second memory part 320 are big or small identical to 325A's.In detail, in these memory blocks, each memory block can comprise address 1 to 2048 (sexadecimal is 0x800).

Similarly, the total value of the total value of the memory block 311B in the first memory part 310, the memory block 312B in the first memory part 310, memory block 312C and 313C and memory block 324B and 325B is big or small identical.Specifically, in these memory blocks, each memory block can comprise address 1 to 768 (sexadecimal is 0x300).

Figure 20 shows the structure of operation part 330.As shown in figure 20, operation part 330 comprises that a subrane signal generation divides the continuous transfer instruction part of 331, subrane composite parts 332,333, virtual address distribution portion 334, subrane signal to add/subtract part 335, subrane signal hop 336, staggered scanning part 337, special data transmission and finishes test section 338 and one and divide decoding part 339.

The subrane signal generation divides the 331 signal interpretation wavelet segment signals with the outside input.In order to reduce the synthetic required calculated amount of subrane, the subrane signal adds/subtracts part 335 and before the subrane synthetic operation subrane signal carried out pre-service (referring to " subrane filtering fast " that Konstantinos Konstantinides delivers, IEEE 1994) in " mpeg audio decoding ".

Subrane composite part 332 utilizes subrane synthetic filtering data and subrane signal to carry out the operation of subrane synthetic filtering by passage ground, produces sound signal.Staggered scanning part 337 is reset according to predetermined order from each passage collection PCM data sample and with the sample of gathering.Staggered scanning part 337 comprises that part 337A is divided in a staggered scanning and part 337B is selected in a staggered scanning memory block.Staggered scanning is divided part 337A the staggered scanning processing is divided into the r step, in this r 〉=2.

In this example, suppose to carry out 2 steps (being first half and latter half) staggered scannings processing of dividing part 337A five equilibrium by staggered scanning.The port number that the staggered scanning memory block selects part 337B to be used for being used for according to input the audio coding signal of subrane synthetic operation decoding is odd number or even number, uses memory block in the first memory part 310 to carry out staggered scanning selectively and handles.In detail, when the port number of subrane synthetic operation decoding is even number, the data storage area of selecting first memory block 311 to handle, and the data storage area of selecting second memory block 312 to handle as the latter half staggered scanning as the first half staggered scanning.When the port number of decoding is odd number, the data storage area of selecting second memory block 312 to handle, and the data storage area of selecting first memory block 311 to handle as the latter half staggered scanning as the first half staggered scanning.

Transfer instruction part 333 specifies in the special memory block in first memory part 310 and the second memory part 320 respectively continuously, and order data transmission control unit (TCU) 340 is carried out one or more data transfer operations.Continuously transfer instruction part 333 comprises that a PCM data transfer instruction part 333A, a PCM data transmission storage area select part 333B and a special data transmission to finish and detect indicating section 333C.

The PCM data of t-2 passage in t the passage of PCM data transfer instruction part 333A order transmission between first memory part 310 and second memory part 320, these data are data of deciphering in first to t-2 subrane synthetic operation respectively, at this, t is the port number (t 〉=3) of input by the sound signal of subrane synthetic operation decoding.In detail, PCM data transfer instruction part 333A order will be respectively in the first PCM data of deciphering in (t-2) individual subrane synthetic operation, subrane synthetic filtering data together with respective channel are transferred to the second memory part 320 from first memory part 310.And PCM data transfer instruction part 333A also orders with the subrane synthetic operation of last passage and mutually concurrently will be respectively be transferred to the first memory part 310 from second memory part 320 in the first PCM data of deciphering in (t-2) individual subrane synthetic operation.

It is odd number or even number that the PCM data storage area selects part 333B to be used for according to the port number by the audio coding signal of subrane synthetic operation decoding, uses the memory block in first memory part 310 selectively, and the PCM data transmission is to this memory block.In these memory blocks, concurrently the PCM data are transferred to the first memory part 310 from second memory part 320 mutually with the subrane synthetic operation of last passage of in PCM data transfer instruction part 333A, carrying out.In this example, when the port number of decoding is even number, select first memory block 311, and when the port number of decoding is odd number, select second memory block 312.

The p that detects in the individual data transfer operation of carrying out of q (q＞1) that indicating section 333C will detect is finished in special data transmission between first memory part 310 and second memory part 320 (the individual transmission of the 1≤p＜q) performance of operating is notified to Data Transmission Control Unit 340, in this example, when order subrane synthetic filtering data that last passage subrane is synthetic when first memory part 310 is transferred to second memory part 320, subrane synthetic filtering data first hop (corresponding to the A2 among Figure 10) of supposing last passage is to be operated by p transmission to transmit.

Only when the synthetic port number of the subrane that is used to import the audio code data be 3 or when bigger, subrane signal hop 336 just is transferred to part or all the specific channel subrane signal in the first memory part 310 in the special section of first memory part 310.

Special data transmission is finished test section 338 and is received p in the individual data transfer operation of carrying out of q (q＞1) that indication detected from Data Transmission Control Unit 340 (1≤p＜q) signal of performance is operated in individual transmission between first memory part 310 and second memory part 320.

There is a virtual address in virtual address distribution portion 334 hypothesis after the actual address of each memory block 311A and 312A end points, and the starting point of virtual address is distributed to each memory block 311A and the initial actual address of 312A.In this manner, virtual address distribution portion 334 is further sequentially distributed to virtual address the actual address of back.

The subrane signal that is used to decipher each passage one frame sound signal is handled in order to carry out, subrane is synthetic handles and staggered scanning is handled, division decoding part 339 is that a signal frame is divided into y processing block (y≤2) with processing unit, and carries out the synthetic processing of subrane.When producing b subrane and c sample in a processing block, the sample number of the PCM data that produce in a signal frame of each passage is b * c * y.In this example, in order to simplify narration, making y=6, is 1 at the processing unit of dividing in handling.

For example, according at the form described in ISO/IEC 11172-3:1993 and the 13818-3:1996, it is 32 that the subrane signal generation divides the partial wave hop count of 331 each the passage subrane signal that produces.The sample number of one frame subrane sample of signal is 1152 in each passage.In this example, the signal frame that the subrane signal is handled because each passage is carried out, subrane synthesizes processing and staggered scanning is handled all is divided into 6 processing blocks, and therefore the sample number of subrane signal that produces in a processing block and PCM data is individual for each passage 192 (32 * 6).

Data Transmission Control Unit shown in Figure 19 340 is according to the instruction of transfer instruction part 333 continuously, data is transferred to the second memory part 320 or from second memory part 320 from first memory part 310 be transferred to the first memory part 310.Data Transmission Control Unit 340 comprises that a special data transmission that provides therein finishes detected transmission part 341.

P in the individual data transfer operation of q (q＞1) that detected transmission part 341 carries out according to the command detection of transfer instruction part 333 is continuously finished in special data transmission between first memory part 310 and second memory part 320 (1≤p＜q) whether individual operation is finished.

Operation part 330 order in memory block 311 or 312 some memory blocks and the second memory part 320 of external processor between carry out data transmission.The data transmission of operation part and back mutually concurrently executable operations handle.When operation part 330 carries out that subranes are synthetic to be handled, utilize (in the memory block 311 or 312) another memory block to carry out the subrane synthetic filtering operation of some passages.And, handle for executable operations, as long as the data transmission district in the first memory part 310 be used for the memory block of storage computation value and do not cover mutually, just can use any one memory block in memory block 311,312 and 313.

Therefore, the transmission of the generation of sound signal and data is parallel processings.For example, when carrying out data transmission between first memory block 311 of first memory part 310 and second memory part 320, operation part 330 uses the data in second memory block 312 of first memory parts 310 to carry out the subrane synthetic operation.When carrying out data transmission between second memory block 312 and second memory part 320, operation part 330 uses the data in first memory block 311 to carry out the subrane synthetic operation.

In example 5, the arrangement of each passage subrane synthetic filtering data of generation and similar with example 4 (Fig. 9 to Figure 11) over time.The virtual address of distributing to first memory part 310 by virtual address distribution portion 334 shown in Figure 20 is also similar with example 4 (Figure 12) with the corresponding relation of corresponding actual address.And, also identical with reference to the operation of figure 9 to Figure 12 narrations with the operation of example 4 explanations, therefore will no longer narrate at this example.

Figure 21 to Figure 32 shows the data transmission scheme according to this example.12 data transmission Fig. 1 to Figure 12 are described in detail as follows.

Audio frequency decoding device 300 with said structure operates as follows.

At first, hyperchannel (5 passages) bit stream when (comprising coding audio signal) as input MPEG2, operation part 330 flows to the subrane signal generation with audio coding signal and divides 331, and the subrane signal generation divides 331 the audio code signal interpretation become the subrane signal of 5 passages.

The transmission diagram 1 of Figure 21 shows the data transmission in first memory part 310.Afterwards, the subrane signal that operation part 330 will be used for the individual passage of i (i=1 in this example) is written to memory block 311B, the subrane signal that will be used for the individual passage of j (at this this example j=2) is written among the 312B of memory block, the subrane signal that will be used for the individual passage of k (k=3 in this example) is written to the part memory block of memory block 311A, and the subrane signal that will be used for the subrane signal of the individual passage of l (l=4 in this example) and be used for the individual passage of m (m=5 in this example) is written to the part memory block of memory block 312A.

Then, operation part 330 requires the subrane signal to add/subtract part 335 to carry out the subrane signal of third channels and add/reducing.Add at the subrane signal of carrying out third channel/reducing after, operation part 330 order subrane signal hops 336 third channel has been carried out add/the first half subrane signal of reducing is transferred among the 312C of memory block, and the latter half signal is transferred among the 313C of memory block.

The transmission diagram 2 of Figure 22 show be used for first, second, the 4th and the subrane signal in five-way road adding/data transmission of reducing.Transfer instruction part 333 is according to the requirement of operation part 330 continuously, and order data transmission control unit (TCU) 340 transmits data in the following manner.

As shown in figure 22, continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for first passage among the memory block 321A of second memory part 320 to the memory block 311A of first memory part 310.

Therefore, the Data Transmission Control Unit 340 subrane signal data that will be used for first passage is transferred to memory block 311A from memory block 321A.When data transmission was finished, Data Transmission Control Unit 340 was finished detected transmission part 341 by the special data transmission, and notifying operation part 330 has been finished the transmission of data.

Then, operation part 330 requires continuous transfer instruction part 333 to carry out data transmission, make simultaneously subrane add/subtract part 335 start first, second, the 4th and the adding/reducing of the subrane signal in five-way road.Therefore, carry out concurrently first, second, the 4th and the subrane signal in five-way road add/reducing in, the subrane synthetic filtering data that operation part 330 will be used for first passage are transferred to memory block 311A from memory block 321A.

The transmission diagram 2 of Figure 23 shows the subrane signal that the is used for first passage data transmission at synthetic operation.Finish first, second, the 4th and the adding of the subrane signal in five-way road/reducing after, when operation part 330 obtained transmitting the notice of finishing, operation part 330 required continuous transfer instruction part 333 according to following method transmission data.

As shown in figure 23, operation part 330 order will be stored in the subrane signal in the 4th and five-way road among the 321A of memory block and the first half subrane signal of third channel is transferred to respectively among memory block 324B, the 325B and 323B of second memory part 320.Then, operation part 330 order will be stored in be used for second channel among the 322A of memory block subrane synthetic filtering data transmission to memory block 312A.

Parallel mutually with these operations, operation part 330 uses the data in first memory block 311 to start the subrane synthetic filtering operation of first passage by subrane composite part 332.At this moment, the same with example 4, the subrane synthetic filtering data in data field 222-1 (Fig. 9) are updated.In other words, to be 0x680 be updated to the data of 0x6ff the address in first memory block 311.After this, according to formula (6) and (7), carry out 16 multiplication-sum operation and ask for 32 sample audio signal.

In order to keep the consistance of window coefficient assignment order, from the sample of latest update, the subrane generated data assignment that is used to operate is given on sample order ground one by one.In detail, at first give corresponding to the i.e. data assignment in data field 201 of 1/16 sample of the subrane synthetic filtering data of latest update, next then give corresponding near on the i.e. data assignment in data field 216 of 1/16 sample of the subrane synthetic filtering data upgraded before operation, the rest may be inferred, give data field 215,214 ..., until 203 and 202 in the sample assignment.

After the subrane synthetic filtering data assignment in giving data field 207 (0xb80 is to 0xbff) in the data field 206 and during (0x400-) subrane synthetic filtering data assignment of back, for the address is set, need carry out the circulation address process to actual address.The circulation address process is operated by following logical multiply and is realized.At this, the address before the conversion is A, and the address B after the conversion is:

B＝{(A-0x400)×(0x7ff)}+0x400

In the formula * presentation logic takes advantage of operation.

Yet in this example, operation part 330 comprises virtual address distribution portion 334, thereby does not need to carry out the circulation address process that comprises the logical multiply operation.Utilizing virtual address 0xc00 can be distributed in actual address 0x400 to 0xfff is subrane synthetic filtering data in data field 206 to 201,216 and 215 to 0x7ff.

Shown in rectangle region 222-2 (Fig. 9), in follow-up subrane synthetic filtering operation, also upgrade the data in the data field 202, and use the subrane synthetic filtering data in the data field 201,216 to 207 and 206 to 202 to carry out multiplication-sum operation, thus the sound signal of obtaining.

And, shown in rectangle region 222-3 (Fig. 9), in follow-up subrane synthetic filtering operation, also upgrade the data in the data field 203, and utilize the subrane synthetic filtering data in data field 202,201,216 to 207 and 206 to 203 to carry out multiplication-sum operation, thereby the sound signal of obtaining.

Shown in rectangle region 222-6 (Fig. 9), in the 6th the subrane synthetic filtering operation before transmitting audio signal, also upgrade the data in the data field 206, and utilize the subrane synthetic filtering data in data field 205 to 201,216 to 207 and 206 to carry out multiplication-sum operation, thereby the sound signal of obtaining.In sum, in carrying out 6 subrane synthetic filterings operations, need not to carry out the circulation address process and can carry out multiplication-sum operation of utilizing the window coefficient.

The transmission diagram 4 of Figure 24 shows the data transmission of third channel latter half subrane signal in first memory part 310.After finishing the subrane synthetic operation of first passage, the third channel latter half subrane signal data that operation part 330 order subrane signal hops 336 will be stored among the memory block 313C of first memory part 310 is transferred in the latter half of memory block 311B.The first passage PCM data that are used for first passage subrane synthetic operation decoding among the 311B of data field and the latter half of third channel subrane signal are carried out addressing, so that make them each other can be not overlapping.

On the other hand, when operation part 330 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation, and when data transmission was finished, the 330 transmission operations of notifying operation part were finished.

When finishing the operation of the data transmission that is used for first passage and subrane synthetic filtering, transfer instruction part 333 is according to the requirement of operation part 330 continuously, and order data transmission control unit (TCU) 340 transmits data in the following manner.

The transmission diagram 5 of Figure 25 shows the data transmission that is used for second channel in the subrane synthetic operation.As shown in figure 25, continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for first passage among the memory block 311A of first memory part 310 to the memory block 312A of second memory part 320.

Then, continuously transfer instruction part 333 order PCM data transfer instruction part 333A will be stored in the PCM data transmission that is used for first passage among the memory block 311B of first memory part 310 to the memory block 321C of second memory part 320.

And continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for third channel among the memory block 323A of second memory part 320 to the memory block 311A of first memory part 310.Continuous transfer instruction part 333 also orders the subrane signal first half that is used for third channel among the memory block 323B that will be stored in second memory part 320 to be transferred to the memory block 311B first half of first memory part 310.

The third channel subrane signal first half that is used for that is transferred among the 311B of memory block is conjointly located successively, and the latter half of the not overlapping subrane signal that is used for third channel that has transmitted and located.Then, subrane composite part 332 utilizes the subrane synthetic filtering operation of the data startup second channel in second memory block 312.

The data transmission that is used for the subrane synthetic filtering data of first passage is carried out between 222-6 (Fig. 9) and 222-7 (Figure 10).Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.By this operation, the address assignment order of subrane synthetic filtering data that is used for the data transmission after the first passage is identical with the transmission of front.

In this stage, utilize second memory block 312, according to operating identical method, carry out the subrane synthetic filtering operation of second channel with carrying out first passage subrane synthetic filtering.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 340 notifying operation parts 330 had been finished data transmission.

The transmission diagram 6 of Figure 26 shows the data transmission that is used for third channel in the subrane synthetic operation.When the data transmission of finishing second channel and the operation of subrane synthetic filtering, operation part 330 requires continuous transfer instruction part 333 to transmit data in the following manner.

Shown in transmission diagram 6 (Figure 26), continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for second channel among the memory block 312A of first memory part 310 to the memory block 322A of second memory part 320.

Afterwards, continuously transfer instruction part 333 order PCM data transfer instruction part 333A will be stored in the PCM data transmission that is used for second channel among the memory block 312B of first memory part 310 to the memory block 322C of second memory part 320.

And continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for four-way among the memory block 324A of second memory part 320 to the memory block 312A of first memory part 310.

Then, transfer instruction part 333 orders the subrane signal that is used for four-way that will be stored among the 324B of memory block to be transferred to the memory block 312B of first memory part 310 continuously.

Subrane composite part 332 utilizes the data in the memory block 311, starts the subrane synthetic filtering operation that is used for third channel concurrently with above-mentioned transmission operation.

The data transmission that is used for second channel subrane synthetic filtering data is to carry out between 222-6 (Fig. 9) and 222-7 (Figure 10) according to the mode identical with first passage.Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.

In this stage, utilize memory block 311, according to operating identical method, carry out the subrane synthetic filtering operation of third channel with the subrane synthetic filtering of execution first or second channel.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 340 notifying operation parts 330 had been finished data transmission.

The transmission diagram 7 of Figure 27 shows the data transmission that is used for four-way in the subrane synthetic operation.When the data transmission of finishing third channel and the operation of subrane synthetic filtering, operation part 330 requires continuous transfer instruction part 333 to transmit data in the following manner.

Shown in transmission diagram 7 (Figure 27), continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for third channel among the memory block 311A of first memory part 310 to the memory block 323A of second memory part 320.

Afterwards, continuously transfer instruction part 333 order PCM data transfer instruction part 333A will be stored in the PCM data transmission that is used for third channel among the memory block 311B of first memory part 310 to the memory block 323C of second memory part 320.

And continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for the five-way road among the memory block 325A of second memory part 320 to the memory block 311A of first memory part 310.

Then, transfer instruction part 333 orders the subrane signal that is used for the five-way road that will be stored among the 325B of memory block to be transferred to the memory block 311B of first memory part 310 continuously.

Subrane composite part 332 utilizes the data in the memory block 312, and the subrane synthetic filtering that starts four-way with aforesaid operations is concurrently operated.

The data transmission that is used for the subrane synthetic filtering data of third channel is according to the mode identical with first or second channel, carries out between 222-6 (Fig. 9) and 222-7 (Figure 10).Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.

In this stage, utilize memory block 312, according to the identical method of subrane synthetic filtering operation of first, second or third channel, carry out the subrane synthetic filtering operation of four-way.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 340 notifying operation parts 330 had been finished data transmission.

The transmission diagram 8 of Figure 28 shows the data transmission in the subrane synthetic operation in five-way road.When finishing the operation of the data transmission that is used for four-way and subrane synthetic filtering, operation part 330 requires continuous transfer instruction part 333 to transmit data in the following manner.

Shown in transmission diagram 8 (Figure 28), transfer instruction part 333 orders the subrane synthetic filtering data transmission of the four-way among the memory block 312A that will be stored in first memory part 310 in the memory block 324A of second memory part 320 continuously.

Afterwards, continuously transfer instruction part 333 is selected part 333B by PCM data transfer instruction part 333A and PCM data storage area, and order will be used for first, second and third channel PCM data and be transferred to first memory part 310 from memory block 321C, 322C and the 323C of second memory part 320.

In this example, because the port number of subrane synthetic operation decoding is an odd number, the PCM data storage area selects part 333B to select the memory block of the memory block afterwards, memory block of the storage four-way PCM data in the memory block 312 as the PCM data.

Subrane composite part 332 utilizes the data in the memory block 311, starts the subrane synthetic filtering operation that is used for the five-way road concurrently with above-mentioned transmission operation.

The data transmission of the subrane synthetic filtering data of four-way is according to the mode identical with first, second or third channel, carries out between 222-6 (Fig. 9) and 222-7 (Figure 10).Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, the data head of the some data fields of transmission transmits the data in other district more earlier.

In this stage, utilize memory block 311, according to first, second, third or the identical method of the subrane synthetic filtering of four-way operation, carry out the subrane synthetic filtering operation in five-way road.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When transmission was finished, Data Transmission Control Unit 340 notifying operation parts 330 had been finished data transmission.

The transmission diagram 9 of Figure 29 shows the data transmission in first half staggered scanning operation.When the data transmission of finishing the five-way road and the operation of subrane synthetic filtering, operation part 330 requires continuous transfer instruction part 333 to transmit data in the following manner.

Shown in transmission diagram 9 (Figure 29), transfer instruction part 333 orders the subrane synthetic filtering data transmission in the five-way road among the memory block 311A that will be stored in first memory part 310 in the memory block 325A of second memory part 320 continuously.

The data transmission of five-way road subrane synthetic filtering data is according to first, second, third or the identical mode of four-way, carries out between 222-6 (Fig. 9) and 222-7 (Figure 10).Rectangle region 200 is divided into A1 and A2 two parts, and A1 comprises data field 206 to 201 and 216 to 213, and A2 comprises data field 212 to 207, and in these data fields, at first transmits the data of some data fields, transmits the data in other district again.

The five-way road is last passage of subrane synthetic operation, and the subrane synthetic filtering data transmission of last passage is carried out in two steps.The special data transmission is finished and is detected finishing of indicating section 333C detection first's transmission shown in Figure 29, and requires Data Transmission Control Unit 340 to carry out data transfer operations.

Staggered scanning part 337 utilizes PCM data that are stored in first, second, third and four-way in the memory block 312 and the PCM data that are stored in the five-way road in the memory block 311 to begin the staggered scanning operation with above-mentioned transmission operation mutually concurrently.

Staggered scanning part 337 is divided part 337A division result according to staggered scanning, carries out the staggered scanning operation in two steps.In detail, at first use preceding 96 samples (being first half) of 192 PCM data samples of each passage to carry out first half staggered scanning operation, use back 96 samples (being latter half) to carry out latter half staggered scanning operation then.

Staggered scanning operation shown in the transmission diagram 9 (Figure 29) is corresponding with the first half staggered scanning operation of dividing.Because the port number of subrane synthetic operation decoding is an odd number in this example, the memory block after the PCM data storage area in the staggered scanning memory block selection part 337B selection memory block 312 is as the memory block of first half staggered scanning data.

When staggered scanning part 337 was carried out first half staggered scanning operation, Data Transmission Control Unit 340 was carried out data transfer operation.When first hop of the subrane synthetic filtering data in five-way road had been transmitted, the finishing of first hop transmission that detected transmission part 341 detects the subrane synthetic filtering data in five-way roads finished in the special data transmission.Then, after the signal that detected transmission part 341 will indicate detection to finish is transferred to operation part 330, start the data transfer operation of second hop of five-way road subrane synthetic filtering data.

Transmission Figure 10 of Figure 30 shows the data transmission in latter half staggered scanning operation.Operation part 330 detects the performance of finishing and being transmitted first hop transmission of the subrane synthetic filtering data of finishing the five-way road of carrying out test section 338 by special data of the first half staggered scanning operation of staggered scanning part 337 execution.Then, operation part 330 is restarted the staggered scanning operation by staggered scanning part 337.

Staggered scanning operation shown in transmission Figure 10 (Figure 30) is corresponding with the latter half staggered scanning operation of dividing.Because in this example, the port number of subrane synthetic operation decoding is an odd number, and the memory block in the staggered scanning memory block selection part 337B selection memory block 312 after the PCM data storage area, five-way road is as the memory block of latter half staggered scanning data.

Transmission Figure 11 of Figure 31 shows the data transmission in first memory part 310.When staggered scanning part 337 is finished latter half staggered scanning when operation, operation part 330 will be stored in latter half staggered scanning data transmission in the memory block 311 of first memory part 310 in memory block 312, as shown in figure 31.

Transmission Figure 12 of Figure 32 shows the data transmission of staggered scanning data.When latter half staggered scanning data had been transferred in the memory block 312, operation part 330 required continuous transfer instruction part 333 to transmit data in the following manner.

Shown in figure 32, continuously 333 orders of transfer instruction part will be stored in the memory block 326 of staggered scanning data transmission in the second memory 320 in the memory block 312 of first memory 310.

The data transmission of staggered scanning data is carried out (being first half and latter half) in two steps.At first transmit first half and latter half the two one of the staggered scanning data, and then the transmission another part data.In the first half of the memory block 326 of first half staggered scanning data storage in second memory part 320, and in the latter half of the memory block 326 of later half part staggered scanning data storage in second memory part 320.

Example 6

The audio frequency decoding device of example 6 receives hyperchannel (4 passages) coded signal that is used for the MPEG2 second layer, and coded signal is decoded into sound signal.In ISO/IEC 11172-3:1993 and 13818-3:1996, the MPEG2 second layer there is detailed narration.

The similar of the structure of example 6 decoding equipments and Figure 19 and example 5 shown in Figure 20, unique difference is the port number difference.Therefore, will no longer be described herein the structure of the decoding equipment of example 6 hereinafter.

In example 6, the arrangement of each passage subrane synthetic filtering data of generation and in this time permutations and example 4 (Fig. 9 to Figure 11) similar.It is also similar to example 4 (Figure 12) to distribute to the virtual address and the corresponding relation between the corresponding actual address of first memory part 310 by virtual address distribution portion 334 in Figure 20.

Below with reference to the operation of Figure 33 to the decoding equipment of data transmission scheme narration example 6 shown in Figure 43.

At first, hyperchannel (4 passages) bit stream when (comprising coding audio signal) as input MPEG2, operation part 330 flows to the subrane signal generation with the sound signal of encoding and divides 331, and the subrane signal generation divides 331 sound signals with coding to be decoded into 4 passage subrane signals.Then, the subrane signal that will be used for the individual passage of i (i=1 in this example) is written to memory block 311B, and the subrane signal that will be used for the individual passage of j (j=2 in this example) is written to memory block 312B.The subrane signal that will be used for the individual passage of k (k=3 in this example) is written to the part memory block of memory block 311A, and the subrane signal that will be used for the individual passage of l (l=4 in this example) is written to the part memory block of memory block 312A.

The transmission diagram 1 of Figure 33 shows the subrane signal that is used for third channel in first memory part 310 and adds/data transmission of reducing.Operation part 330 requires the subrane signal to add/subtract part 335 and carries out adding/reducing of third channel subrane signal.Then, after the adding of third channel subrane signal/reducing is finished, operation part 330 order will carry out add/the subrane signal first half of the third channel of reducing is transferred among the 312C of memory block, the latter half signal is transferred among the 313C of memory block.

The transmission diagram 2 of Figure 34 shows the subrane signal that is used for first, second and four-way and adding/data transmission of reducing.Transfer instruction part 333 is transmitted data in the following manner according to the order data transmission control unit (TCU) 340 that requires of operation part 330 continuously.

As shown in figure 34, continuously 333 orders of transfer instruction part will be stored in and be used for the memory block 311A of first passage subrane synthetic filtering data transmission to first memory part 310 among the memory block 321A of second memory part 320.

Therefore, the Data Transmission Control Unit 340 subrane signal data that will be used for first passage is transferred to memory block 311A from memory block 321A.When data transmission was finished, Data Transmission Control Unit 340 notifying operation parts 330 had been finished the transmission of data.

Make the subrane signal add/subtract the subrane signal that part 335 starts first, second and four-way to add/reducing in, operation part 330 further requires continuous transfer instruction part 333 to carry out data transmission.Thereby, add at the subrane signal of carrying out first, second and four-way concurrently/reducing in, the subrane synthetic filtering data that will be used for first passage are transferred to memory block 311A from memory block 321A.

The transmission diagram 3 of Figure 35 shows the subrane signal that the is used for first passage data transmission at synthetic operation.After the adding of the subrane signal of finishing first, second and four-way/reducing, when operation part 330 obtained transmitting the notice of finishing, operation part 330 required continuous transfer instruction part 333 according to following method transmission data.

As shown in figure 35, operation part 330 order will be stored among the 321A of memory block be used for second and the subrane signal of four-way and the first half subrane signal that is used for third channel be transferred to memory block 323B, 324B and the 325B of second memory part 320 respectively.Then, operation part 330 order will be stored in be used for second channel among the 322A of memory block subrane synthetic filtering data transmission to memory block 312A.

When being parallel to these operations, operation part 330 uses the data in first memory block 311 to start the subrane synthetic filtering operation of first passage by subrane composite part 332.

The transmission diagram 4 of Figure 36 shows the data transmission of the latter half subrane signal that is used for third channel in first memory part 310.After the subrane synthetic operation of first passage is finished, operation part 330 order subrane signal hops 336 will be stored in the latter half memory block of the latter half data transmission of the subrane signal that is used for third channel among the memory block 313C of first memory part 310 to memory block 311B.Carry out addressing with being used for the first passage PCM data of subrane synthetic operation decoding of first passage and the latter half of third channel subrane signal among the 311B of data field, so that make them not overlapped.

The transmission diagram 5 of Figure 37 shows the data transmission in second channel subrane synthetic operation.When operation part 330 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation, and when the transmission operation was finished, notifying operation part 330 had been finished data transmission.

When the data transmission of first passage and the operation of subrane synthetic filtering had all been finished, transfer instruction part 333 required order data transmission control unit (TCU) 340 according to following method transmission data according to operation part 330 continuously.

As shown in figure 37, continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for first passage among the memory block 311A of first memory part 310 to the memory block 321A of second memory part 320.

Afterwards, continuously transfer instruction part 333 order PCM data transfer instruction part 333A will be stored in the PCM data transmission that is used for first passage among the memory block 311B of first memory part 310 to the memory block 321C of second memory part 320.

And continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for third channel among the memory block 323A of second memory part 320 to the memory block 311A of first memory part 310.Transfer instruction part 333 also orders the first half of the subrane signal that is used for third channel among the memory block 323B that will be stored in second memory part 320 to be transferred to the first half of the memory block 311B of first memory part 310 continuously.

Sequentially the first half that is transferred to the third channel subrane signal among the 311B of memory block is positioned, make it not cover the latter half of the third channel subrane signal that has transmitted and located.Then, subrane composite part 332 utilizes the subrane synthetic filtering operation of the data startup second channel in second memory block 312.

In this step,, utilize second memory block 312 to carry out the subrane synthetic filtering operation of second channels according to operating identical method with carrying out first passage subrane synthetic filtering.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When having finished data transmission, Data Transmission Control Unit 340 notifying operation parts 330 data transmission are finished.

The transmission diagram 6 of Figure 38 shows the data transmission in third channel subrane synthetic operation.When the data transmission of second channel and the operation of subrane synthetic filtering had all been finished, operation part 330 required continuous transfer instruction part 333 according to following method transmission data.

Shown in transmission diagram 6 (Figure 38), continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for second channel among the memory block 312A of first memory part 310 to the memory block 322A of second memory part 320.

Afterwards, continuously transfer instruction part 333 order PCM data transfer instruction part 333A will be stored in the PCM data transmission that is used for second channel among the memory block 312B of first memory part 310 to the memory block 322C of second memory part 320.

And continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for four-way among the memory block 324A of second memory part 320 to the memory block 312A of first memory part 310.

Then, transfer instruction part 333 orders the subrane signal that is used for four-way that will be stored among the 324B of memory block to be transferred to the memory block 312B of first memory part 310 continuously.

Subrane composite part 332 utilizes data and the above-mentioned transmission operation in the memory block 311 to start the subrane synthetic filtering operation that is used for third channel concurrently.

In this step, according to carry out first or the identical method of operation of second channel, utilize memory block 311 to carry out the subrane synthetic filtering operation of third channels.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When finishing data transmission, Data Transmission Control Unit 340 notifying operation parts 330 data transmission are finished.

The transmission diagram 7 of Figure 39 shows the data transmission in four-way subrane synthetic operation.When the data transmission of third channel and the operation of subrane synthetic filtering had all been finished, operation part 330 required continuous transfer instruction part 333 according to following method transmission data.

Shown in transmission diagram 7 (Figure 39), continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for third channel among the memory block 311A of first memory part 310 to the memory block 323A of second memory part 320.

Afterwards, continuous transfer instruction part 333, select part 333B by PCM data transfer instruction part 333A and PCM data storage area, the PCM data that order will be used for first and second passages are transferred to first memory part 310 from the memory block 321C and the 322C of second memory part 320.

In this example, because the port number of subrane synthetic operation decoding is an even number, therefore, the PCM data storage area selects part 333B to be chosen in the memory block afterwards, PCM data storage area that being used in the memory block 311 stored third channel.

Subrane composite part 332 utilizes data and the above-mentioned transmission operation in the memory block 312 to start the subrane synthetic filtering operation that is used for four-way concurrently.

In this step, according to the identical method of operation of carrying out first, second or third channel, utilize memory block 311 to carry out the subrane synthetic filtering operation of four-ways.When subrane composite part 332 was carried out the operation of subrane synthetic filtering, Data Transmission Control Unit 340 was carried out data transfer operation.When finishing data transmission, Data Transmission Control Unit 340 notifying operation parts 330 data transmission are finished.

The transmission diagram 8 of Figure 40 shows the data transmission in first half staggered scanning operation.When the data transmission of having finished four-way and the operation of subrane synthetic filtering, operation part 330 requires continuous transfer instruction part 333 to transmit data in the following manner.

Shown in transmission diagram 8 (Figure 40), continuously 333 orders of transfer instruction part will be stored in the subrane synthetic filtering data transmission that is used for four-way among the memory block 312A of first memory part 310 to the memory block 324A of second memory part 320.

The data transmission that is used for the subrane synthetic filtering data of four-way is to carry out according to the mode identical with first, second or third channel, meanwhile, data field shown in the 222-6 (Fig. 9) is divided into A1 and A2 two parts, A1 comprises data field 206 to 201 and 216 to 213, A2 comprises data field 212 to 207, and in these data fields, at first transmit the data of some data fields, transmit the data in other district again.

Four-way is last passage of subrane synthetic operation, and the subrane synthetic filtering data transmission of last passage in two steps.The special data transmission is finished and is detected indicating section 333C detection finishing that the first shown in Figure 29 transmits, and requires Data Transmission Control Unit 340 to carry out data transfer operations.

The staggered scanning that staggered scanning part 337 and above-mentioned transmission operation start the PCM data that are stored in first, second and third channel in the memory block 311 mutually concurrently and be stored in the PCM data of the four-way in the memory block 312 is operated.

In this example, staggered scanning division part 337A became for two steps with the staggered scanning division of operations of staggered scanning part 337 execution.In detail, at first use preceding 96 samples (being first half) of 192 PCM data samples of each passage to carry out first half staggered scanning operation, use back 96 sampling points (being latter half) to carry out latter half staggered scanning operation then.

Staggered scanning operation shown in the transmission diagram 8 (Figure 40) is corresponding with the first half staggered scanning operation of dividing.Because the port number of subrane synthetic operation decoding is an even number in this example, the memory block in the staggered scanning memory block selection part 337B selection memory block 311 after the PCM data storage area is as the memory block of storage first half staggered scanning data.

Data Transmission Control Unit 340 is carried out data transfer operation when staggered scanning part 337 is carried out first half staggered scanning operation.When first hop of the subrane synthetic filtering data of four-way had been transmitted, the finishing of first hop transmission that detected transmission part 341 detects the subrane synthetic filtering data of four-ways finished in the special data transmission.Then, after the signal that detected transmission part 341 will indicate detected transmission to finish is transferred to operation part 330, start the data transfer operation of second hop of four-way subrane synthetic filtering data.

The transmission diagram 9 of Figure 41 shows the data transmission in latter half staggered scanning operation.Operation part 330 detects the performance of finishing and being finished first hop transmission of the four-way subrane synthetic filtering data of carrying out test section 338 by the special data transmission of the first half staggered scanning operation of staggered scanning part 337 execution.Then, operation part 330 is restarted the staggered scanning operation by staggered scanning part 337.

Staggered scanning operation shown in the transmission diagram 9 (Figure 41) is corresponding to the latter half staggered scanning operation of dividing.In this example, because the port number of subrane synthetic operation decoding is an even number, the memory block in the staggered scanning memory block selection part 337B selection memory block 312 after the four-way PCM data storage area is as the memory block of storage latter half staggered scanning data.

Transmission Figure 10 of Figure 42 shows and is used to finish the interleaved data transmission of latter half.When having finished the latter half staggered scanning operation of data transmission and 337 execution of staggered scanning part, operation part 330 will be stored in first half staggered scanning data transmission in the memory block 311 of first memory part 310 in memory block 312, as shown in figure 31.

Transmission Figure 11 of Figure 43 shows the data transmission of staggered scanning data in first memory part 310.When latter half staggered scanning data had been transferred in the memory block 312, operation part 330 required continuous transfer instruction part 333 to transmit data in the following manner.

As shown in figure 43, continuously 333 orders of transfer instruction part will be stored in staggered scanning data transmission in the memory block 312 of first memory 310 in the memory block 326 of second memory 320.

The data transmission of staggered scanning data is carried out (being first half and latter half) in two steps.At first transmit the staggered scanning data of first half or latter half, and then the data of transmission another part.First half staggered scanning data storage is in the first half of the memory block 326 of second memory part 320, and later half part staggered scanning data storage is in the latter half of the memory block 326 of second memory part 320.

After this, export the decoding pcm audio signal of four passages simultaneously.

In the decoding equipment of example 5 and example 6, each decoding equipment all used one than the cheap DRAM of SRAM as second memory part 320.Be stored in the data of all passages in the second memory part 320 and other data in 320, all can use Data Transmission Control Unit 340 to be transferred in the first memory part 310 as required.Therefore, need not to enlarge the internal memory of processor and the coding audio signal of a plurality of passages of can regenerating at high speed.And, can also remove external SRAM, thereby reduce the cost of equipment.

And in the decoding equipment of example 5 and example 6, each decoding equipment all can be carried out the operation of subrane synthetic filtering mutually concurrently with data transmission, therefore can shorten the required time of operational processes by the required time of data transmission.Because the data volume corresponding to a plurality of transmission operations can be transmitted by a transmission operation, therefore, in the subrane synthetic operation that has used the virtual address distribution portion, does not need to carry out the circulation address process and the shifting processing of subrane synthetic filtering data.

And in the decoding equipment of each example 5 and example 6, the subrane signal of k passage is temporarily to be transferred to first memory part 310, has therefore shortened the required time of data transmission.

And, in the decoding equipment of each example 5 and example 6, the adding of subrane signal/reducing is to carry out concurrently mutually from the operation that second memory part 320 is transferred to the first memory part 310 with the subrane synthetic filtering data of i passage, thereby has shortened the stand-by period of data transmission.

And, in the decoding equipment of each example 5 and example 6, the adding of the subrane signal of i, j, k, l and m passage/reducing is to carry out concurrently mutually from the operation that second memory part 320 is transferred to the first memory part 310 with i passage subrane synthetic filtering data, thereby has shortened the stand-by period of data transmission.

And, in the decoding equipment of each example 5 and example 6, after in the end the subrane synthetic operation of a passage is finished, the staggered scanning part is carried out the staggered scanning operation with subrane synthetic filtering data mutually concurrently from the transmission operation that first memory partly is transferred to the second memory part, thereby has reduced the stand-by period of data transmission.

And in the decoding equipment of example 5 and example 6, each decoding equipment includes staggered scanning and divides part, is used for the staggered scanning division of operations is become the r step (r=2 in above-mentioned example).Therefore, because the staggered scanning operation can be divided into a plurality of operation steps, so do not need to provide the connected storage that is used for storing the staggered scanning data.Correspondingly reduce the size of first memory part, thereby reduced the cost of equipment.And the division operation of this staggered scanning operation has reduced to carry out a required continuous time in operation step, thereby has simplified the operational design parallel with data transfer operation.

And, in the decoding equipment of each example 5 and example 6, can detect the special data transmission operation in a series of data transfer operations, therefore,, can arrange signal Processing efficiently by the staggered scanning operation of dividing.

And in the decoding equipment of example 5 and example 6, each decoding equipment includes PCM data transfer instruction part, be used for the subrane synthetic operation mutually concurrently temporarily with the PCM data transmission in second memory part.Therefore, only need in the end the subrane synthetic operation of a passage to operate in this time, guarantee to be used for storing the memory block of all passage PCM data that need to carry out the staggered scanning operation to staggered scanning.The subrane synthetic filtering data transmission of passage arrives after the second memory part in front, and the PCM data that have been transmitted and temporarily have been stored in the second memory part can be distributed in the subrane signal that has stored the front passage in the first memory part and the memory block of subrane synthetic filtering data.Therefore, do not need to be provided for storing the special memory block of PCM data, thereby reduced the size of first memory part in the first memory part.

And in the decoding equipment of example 5 and example 6, each decoding equipment includes the PCM data storage area and selects part.Therefore, when the PCM data that will need to carry out the staggered scanning operation when second memory part 320 is transferred to the first memory part 310, the port number no matter the subrane synthetic operation is deciphered is odd number or even number, and data transfer operation can be carried out concurrently with last passage subrane synthetic operation.Therefore, can correspondingly reduce required calculated amount.

And, in the decoding equipment of example 5 and example 6, each decoding equipment includes the staggered scanning memory block and selects part, the port number no matter the subrane synthetic operation is deciphered is odd number or even number, can not utilize not the memory areas that covers mutually with the data transmission district that is used for last passage subrane synthetic operation, start the staggered scanning operation.Therefore, staggered scanning operation can with the data transfer operation executed in parallel.

And, in the decoding equipment of example 5 and example 6, each decoding equipment includes divides the decoding part, is used for that a signal frame is divided into 6 parts and is used for decoded operation, thereby only need store the subrane signal of each passage corresponding to the memory block of 1/6 signal frame size.Therefore, correspondingly reduced the size of first memory part.

In example 5 and example 6,, suppose that the port number of input signal is respectively 5 and 4 in order to simplify narration.Yet port number is not limited only to above-mentioned particular value.

And, in the decoding equipment of each example 5 and example 6,, suppose that it is 2 that the staggered scanning operation step number of part division is divided in staggered scanning in order to simplify narration.Yet the staggered scanning operation can be divided into 3 step or multisteps, and does not need five equilibrium.

And, in the decoding equipment of each example 5 and example 6,, suppose that memory block 312 and 313 is used as the terminal point of subrane signal hop transmission subrane signal in order to simplify narration.Yet the terminal point of data transmission is not limited only to above-mentioned memory block.And the quantity and the size of data that are used to transmit the subrane division of signal of operation are not limited only to above-mentioned value.

In the decoding equipment of each example 5 and example 6, when the port number of subrane synthetic operation decoding is even number, select first memory block 311 to store the first half of staggered scanning data, select second memory block 312 to store the latter half of staggered scanning data simultaneously.When the port number of subrane synthetic operation decoding is odd number, select second memory block 312 to store the first half of staggered scanning data, select first memory block 311 to store the latter half of staggered scanning data simultaneously.Yet this selection is not limited thereto, and it can change according to the arrangement of operation and data transmission.

And, in the decoding equipment of each example 5 and example 6, in order to simplify narration, suppose when the port number of subrane synthetic operation decoding is even number, the PCM data storage area selects part 333B to select first memory block 311, and when the port number of subrane synthetic operation decoding was odd number, the PCM data storage area selected part 333B to select second memory block 312.Yet this selection is not limited thereto, and it can change according to the arrangement of operation and data transmission.

Above-mentioned example 4 to the compressed code decoding device of example 6 has following effect.

Can reduce the capacity of required high speed access storage.Therefore, compare the cost that can correspondingly reduce decoding equipment with conventional equipment.

The operation of subrane synthetic filtering can be carried out mutually concurrently with data transfer operation.Thereby can fully save the required time of data transmission, thereby shorten the processing time widely.

Issuing an instruction can a plurality of data transfer operation of command execution, therefore, can utilize predetermined address to divide the data that will be transmitted, then transmission data of dividing one by one.Thereby, can shorten the required time of circulation address process.And, owing to can carry out a plurality of data transfer operations, and therefore, according to producing the required time of output data, the operation of arranging data transmission at an easy rate.For example, when carrying out a plurality of data transfer operation concurrently with the subrane synthetic operation, do not need to interrupt the subrane synthetic operation wait for or confirm data transmission finish or issue transmission command.Therefore, can reduce the time of data transmission loss, thereby shorten the time of entire process.

Because the virtual address distribution portion is provided, does not therefore need to carry out the required operation of circulation address process, thereby shorten the time of entire process.

Provide to be used for temporarily third channel subrane signal being transferred to the subrane signal hop of first memory part, thereby reduced required volume of transmitted data.Correspondingly shorten the required time of data transmission, therefore reduced required calculated amount.

The subrane signal that is independent of subrane composite part hop is provided.Therefore, after the coding audio data of input is decoded into the subrane signal of different subranes, can carry out adding/reducing of subrane signal mutually concurrently from first data transfer operation that second memory partly is transferred to the first memory part with subrane synthetic filtering data.Therefore, shorten the stand-by period of data transmission, thereby correspondingly reduced required calculated amount.

The staggered scanning part is provided, has been used for after in the end the subrane synthetic operation of a passage is finished, partly be transferred to second memory partial data transmission operation with subrane synthetic filtering data from first memory and carry out staggered scanning mutually concurrently and operate.Therefore, shorten the stand-by period of data transmission, thereby correspondingly reduced required calculated amount.

Owing to the staggered scanning division part that is used for dividing the staggered scanning operation is provided, has not therefore needed to provide the connected storage that is used for storing the staggered scanning data.Thereby correspondingly reduced the size of the storer of required first memory part, thereby reduced the cost of equipment.And the division of this staggered scanning operation is handled to have shortened and is carried out a required continuous time in operation step, thereby has accelerated the operational design that will carry out parallel mutually with data transfer operation.

Can detect the special data transmission operation in a series of data transfer operations, thereby utilize the staggered scanning operation of dividing, can arrange signal Processing efficiently.

Because being provided, the staggered scanning memory block selects part, thereby, no matter the port number of subrane synthetic operation decoding is odd number or even number, can use the nonoverlapping memory block, data transmission district with last passage subrane synthetic operation to start the staggered scanning operation.Therefore, staggered scanning operation can with the data transfer operation executed in parallel.Thereby correspondingly reduced required calculated amount.

PCM data transfer instruction part and subrane synthetic operation mutually concurrently temporarily with the PCM data transmission to second memory partly in.Therefore, only need in the end the subrane synthetic operation of a passage to operate in this time, guarantee to be used for storing the memory block of all passage PCM data that need to carry out the staggered scanning operation to staggered scanning.The subrane synthetic filtering data transmission of passage arrives after the second memory part in front, and the PCM data that have been transmitted and temporarily have been stored in the second memory part can be distributed in the subrane signal that has stored the front passage in the first memory part and the memory block of subrane synthetic filtering data.Therefore, do not need to be provided for storing the special memory block of PCM data, thereby reduced the size of first memory part in the first memory part.

Because being provided, the PCM data storage area selects part, so, when the PCM data that will need to carry out the staggered scanning operation when second memory partly is transferred to the first memory part, the passage no matter the subrane synthetic operation is deciphered is odd number or even number, can carry out data transfer operation concurrently with the subrane synthetic operation of last passage.Therefore can correspondingly reduce required calculated amount.

Owing to provide decoding to divide part, can reduce to store each required memory block of passage subrane signal, thereby correspondingly reduce the memory size of first memory part.

Any product and various other modification that does not break away from the scope of the invention and spirit all belongs to category of the present invention.Therefore, the scope of the claim that this paper is appended is not limited to scope as herein described, and comprises the more wide scope that claim is explained.

Claims (29)

1. an audio frequency decoding device is used for utilizing subrane synthetic filtering data and subrane signal data, by subrane synthetic operation decoding N _A(N _A＞1) sound signal of individual passage, this equipment comprises:

A first memory part is used for store M _A(M _A＜N _A) the subrane synthetic filtering data that are used for the subrane synthetic operation and the subrane signal data of individual passage;

A second memory part is used for storing N _AThe subrane signal data and the N of individual passage _AAt least some data in the subrane synthetic filtering data of individual passage;

An operation part is used for the voice data of received code and the voice data of coding is decoded into the subrane signal data, and utilizes the data be stored in the first memory part to carry out the operation of subrane synthetic filtering, thus output M _AThe decoding audio data of individual passage, and require to switch position by the new subrane synthetic filtering data of subrane synthetic filtering operational computations and required next subrane synthetic filtering data;

A tcp data segment is used for requirement according to operation part, presses M _AIndividual passage switches subrane synthetic filtering data and the subrane signal data that is stored in first memory part and the second memory part.

2. according to the audio frequency decoding device of claim 1, in this equipment:

First memory comprises that partly first memory block of subrane synthetic filtering data that are used for storing some passages and subrane signal data and one are used for storing the subrane synthetic filtering data of another passage and second memory block of subrane signal data;

When use is stored in data in first memory block of first memory part when carrying out the subrane synthetic filtering operation of i passage, operation part will be stored in the j (i=1 to N of second memory in partly _A, j=1 to N _A, the data transmission of the individual passage of j ≠ i) is in second memory block of first memory part; And when using the data that are stored in first memory second memory block partly to carry out the subrane synthetic filtering operation of j passage, operation part will be stored in the k (k=1 to N in the second memory part _A, k ≠ i, the data transmission of the individual passage of k ≠ j) in first memory block of first memory part, thereby carry out data transfer operation and the operation of subrane synthetic filtering concurrently.

3. according to the audio frequency decoding device of claim 1, operation part comprises a continuous transfer instruction part, be used for when order is carried out data transfer operation between first memory part and second memory part, coming a plurality of data transfer operations of order with an instruction.

4. according to the audio frequency decoding device of claim 3, operation part comprises a virtual address distribution portion, be used for after the actual address of the first memory memory block terminal point partly that is positioned at storage subrane synthetic filtering data, providing virtual address, the starting point of virtual address is distributed to actual address predetermined in the memory block, and sequentially other virtual address is distributed to actual address.

5. according to the audio frequency decoding device of claim 1, in this equipment:

In the first memory part, except that first memory block and second memory block, comprise that also is used for storing unrestricted the 3rd memory block in the data that are used for the subrane synthetic operation;

Operation part comprises a subrane signal hop, being used for port number when the input coding voice data is 3 or greater than 3 the time, at least a portion is stored in the first memory part special modality subrane signal copy or be transferred in the special section of first memory part.

6. according to the audio frequency decoding device of claim 1, operation part comprises that a subrane signal adds/subtract part, after being used for being decoded into the subrane signal data, carry out the addition and the subtraction of subrane signal mutually concurrently with the transmission operation that subrane synthetic filtering data partly are transferred to the first memory part from second memory at the voice data of having imported coding and with it.

7. according to the audio frequency decoding device of claim 1, operation part comprises a staggered scanning part, be used for after in the end the subrane synthetic operation of a passage is finished, gather the decoding data sample of subrane synthetic operation decoding with subrane synthetic filtering data mutually concurrently from the transmission operation that first memory partly is transferred to the second memory part, each passage is gathered a sample, and according to predetermined order the sample of gathering is reset.

8. according to the audio frequency decoding device of claim 7, staggered scanning partly comprises a staggered scanning division part, is used for the staggered scanning division of operations is become the r step (r 〉=2).

9. according to the audio frequency decoding device of claim 7, staggered scanning partly comprises a staggered scanning memory block selection part, be used in the first memory part, selecting to be used for the data storage area of staggered scanning operation according to being odd number or even number by the port number of subrane synthetic operation decoding.

10. according to the audio frequency decoding device of claim 3, continuously transfer instruction comprises that partly a special data transmission finishes the detection indicating section, be used to refer to the performance (q＞1,1≤p＜q) of p data transfer operation in q the data transfer operation of between first memory part and second memory part, carrying out of detection.

11. audio frequency decoding device according to claim 10, tcp data segment comprises that the transmission of special data finishes the detected transmission part, be used for according to the instruction partly of continuous transfer instruction, whether p data transfer operation in q the data transfer operation that detection is carried out between first memory part and second memory part is finished, and p the performance of transmitting operation that is used for detecting is transferred to operation part.

12. according to the audio frequency decoding device of claim 3, in this equipment:

Operation part comprises that the transmission of special data finishes the test section, be used for according to the instruction partly of continuous transfer instruction, detect the performance of p data transfer operation in q the data transfer operation of between first memory part and second memory part, carrying out;

Operation part is finished after the data transfer operation that has detected special section in the test section finishes in special data transmission, carries out s step operation in the r step staggered scanning operation (r 〉=2,2≤s≤r).

13. according to the audio frequency decoding device of claim 3, in this equipment:

Transfer instruction partly comprises a PCM data transfer instruction part continuously, be used for when the port number of importing the decoding data of being deciphered by the subrane synthetic operation is t (t 〉=3), order at least one passage of execution between first memory part and second memory part is deciphered the transmission operation of PCM data;

PCM data transfer instruction part partly is transferred to the second memory part from first memory with the PCM data that the subrane synthetic operation of last passage temporarily will be carried out the subrane synthetic operation mutually concurrently, and the PCM data that will be transferred to the second memory part be transferred to again first memory partly in.

14. audio frequency decoding device according to claim 3, transfer instruction partly comprises a PCM data storage area selection part continuously, being used for according to the passage of importing the decoding data of being deciphered by the subrane synthetic operation is odd number or even number, partly select a memory block at first memory, be used for concurrently the PCM data partly being transferred to the first memory part from second memory mutually with the subrane synthetic operation of last passage.

15. according to the audio code decoding equipment of claim 1, in this equipment:

Operation part comprises that is divided a decoding part, be used for dividing and occur to the decoding processing that this process takes place output signal from the subrane signal, or divide and to be synthesized to the decoding that this process takes place output signal from subrane and to handle, thereby a plurality of audio output signal samples of each frame are divided into y block; And

A=b * c * y, a represents the sample number of audio output signal sample of a frame coding audio signal of each passage, the sample number that the partial wave hop count of b presentation code sound signal, c produce when representing to handle a block in the formula.

16. an audio frequency decoding device is used for utilizing subrane synthetic filtering data and subrane signal data, by subrane synthetic operation decoding N _A(N _A＞1) sound signal of individual passage, this equipment comprises:

A first memory part is used for storing the subrane synthetic filtering data that are used for the subrane synthetic operation and the subrane signal data of at least one passage;

A second memory part is used for storing subrane signal data and N _AThe subrane synthetic filtering data of individual passage;

An operation part, be used for received code voice data and with the coding voice data be decoded into the subrane signal data, and utilize the data that are stored in the first memory part to carry out the operation of subrane synthetic filtering, thereby export the decoding audio data of a passage, and require to switch position by the new subrane synthetic filtering data of subrane synthetic filtering operational computations and required next subrane synthetic filtering data;

A tcp data segment is used for requirement according to operation part, and passage ground switches and is stored in first memory part and second memory subrane synthetic filtering data and the subrane signal data in partly one by one.

17. audio frequency decoding device according to claim 16, operation part comprises a continuous transfer instruction part, be used for when order is carried out data transfer operation between first memory part and second memory part, coming a plurality of data transfer operations of order with an instruction.

18. audio frequency decoding device according to claim 17, operation part comprises a virtual address distribution portion, be used for after the actual address of the first memory memory block terminal point partly that is positioned at storage subrane synthetic filtering data, providing virtual address, and the starting point of virtual address distributed in the memory block predetermined actual address, and sequentially other virtual address is distributed to actual address.

19. according to the audio frequency decoding device of claim 16, in this equipment:

In the first memory part, except that first memory block and second memory block, comprise that also is used for storing unrestricted the 3rd memory block in the data that are used for the subrane synthetic operation;

Operation part comprises a subrane signal hop, being used for port number when the input coding voice data is 3 or greater than 3 the time, at least a portion is stored in the special modality subrane signal copy of first memory part or is transferred in the special section of first memory part.

20. audio frequency decoding device according to claim 16, operation part comprises that a subrane signal adds/subtract part, after being used for being decoded into the subrane signal data, carry out the addition and the subtraction of subrane signal mutually concurrently with the transmission operation that subrane synthetic filtering data partly are transferred to the first memory part from second memory at the voice data of having imported coding and with it.

21. audio frequency decoding device according to claim 16, operation part comprises a staggered scanning part, be used for after in the end the subrane synthetic operation of a passage is finished, gather the decoding data sample of subrane synthetic operation decoding with subrane synthetic filtering data mutually concurrently from the transmission operation that first memory partly is transferred to the second memory part, each passage is gathered a sample, and according to predetermined order the sample of gathering is reset.

22. according to the audio frequency decoding device of claim 21, staggered scanning partly comprises a staggered scanning division part, is used for the staggered scanning division of operations is become the r step (r 〉=2).

23. the audio code according to claim 21 is translated equipment, staggered scanning partly comprises a staggered scanning memory block selection part, be used in the first memory part, selecting to be used for the data storage area of staggered scanning operation according to being odd number or even number by the port number of subrane synthetic operation decoding.

24. audio frequency decoding device according to claim 17, continuously transfer instruction comprises that partly a special data transmission finishes the detection indicating section, be used to refer to the performance (q＞1,1≤p＜q) of p data transfer operation in q the data transfer operation of between first memory part and second memory part, carrying out of detection.

25. audio frequency decoding device according to claim 24, tcp data segment comprises that the transmission of special data finishes the detected transmission part, be used for according to the instruction partly of continuous transfer instruction, the performance of p data transfer operation in q the data transfer operation that detection is carried out between first memory part and second memory part, and p the performance of transmitting operation that is used for detecting is transferred to operation part.

26. according to the audio frequency decoding device of claim 17, in this equipment:

Operation part comprises that the transmission of special data finishes the test section, be used for according to the instruction partly of continuous transfer instruction, detect the performance of p data transfer operation in q the data transfer operation of between first memory part and second memory part, carrying out;

Operation part is finished after the data transfer operation that has detected special section in the test section finishes in special data transmission, carries out s step operation in the r step staggered scanning operation (r 〉=2,2≤s≤r).

27. according to the audio frequency decoding device of claim 17, in this equipment:

Transfer instruction partly comprises a PCM data transfer instruction part continuously, be used for when the port number of importing the decoding data of being deciphered by the subrane synthetic operation is t (t 〉=3), order at least one passage of execution between first memory part and second memory part is deciphered the transmission operation of PCM data;

PCM data transfer instruction part partly is transferred to the second memory part from first memory with the PCM data that the subrane synthetic operation of last passage temporarily will be carried out the subrane synthetic operation mutually concurrently, and the PCM data that will be transferred to the second memory part be transferred to again first memory partly in.

28. audio code decoding equipment according to claim 17, transfer instruction partly comprises a PCM data storage area selection part continuously, being used for according to the passage of importing the decoding data of being deciphered by the subrane synthetic operation is odd number or even number, partly select a memory block at first memory, be used for concurrently the PCM data being operated from the transmission that second memory partly is transferred to the first memory part mutually with the subrane synthetic operation of last passage.

29. according to the audio frequency decoding device of claim 16, in this equipment:

Operation part comprises that is divided a decoding part, be used for dividing and occur to the decoding processing that this process takes place output signal from the subrane signal, or divide and to be synthesized to the decoding that this process takes place output signal from subrane and to handle, thereby a plurality of audio output signal samples of each frame are divided into y block; And

A=b * c * y, a represents the sample number of audio output signal sample of a frame coding audio signal of each passage, the sample number that the partial wave hop count of b presentation code sound signal, c produce when representing to handle a block in the formula.

Images